Scribd Logo
Upload

Welcome to Scribd!

  • CK1 Booklet 1 PDF

    Uploaded by

    Anderson
    73 views89 pages

    Original Title

    CK1_Booklet_1 (16).pdf

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    Download as pdf
    73 views89 pages

    CK1 Booklet 1 PDF

    Uploaded by

    Anderson

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    Download as pdf
    of 89
    Subject CM2 Revision Notes For the 2019 exams Run-off triangles Booklet 9 covering Chapter 21 Run-off triangles The Actuarial Education Company CONTENTS Contents, Page Links to the Course Notes and Syilabus Overviow Core Reading Past Exam Questions Solutions to Past Exam Questions Factsheet fEVaov Copyright agreement All of this matertal is copyright. The copyright belongs to Institule and Facuity Education Ld, a subsidiary ofthe Institute and Faculty of Actuaries. “The materia is sold to you for your own exclusive use. You may not hire cout, lend, give, sell, transmit electronically, store electronically or photocopy any part oft. You must take care of your malerial to ensure itis not used or ccopiad by anyone at any time. Legal action willbe taken if these torms are infinged. In addition, we may seek to take disciplinary action through the profession or through your ‘employer. “These conditions remain in force after you have finished using the course. @IFE: 2019 Examinations Page t LINKS TO THE COURSE NOTES AND SYLLABUS Chapter 21 Run-off angles “This chapter number refers fo the 2019 edition of the AclEd Course Notes. ‘Syllabus objectives covered in this booklet 6, Liability valuations 52 — Run-offtiancles (Chapter 24) 5.2.1 Define a development factor and show how a set of assumed development factors can be used to project the future developmentof a delay triangle. 5.22 Describe and apply a basic chain ladder method for ‘completing the delay triangle using development factors. 5.2.3 Show how the basic chain iadder method can be adjusted to ‘make explicit allowance for inflation. 5.24 Describe and apply the average cost per claim method for ‘estimating oulstanding claim amounts. 5.25 Describe and apply the BornhustterFerquson method for estimating outstanding claim amounts, 5.26 Deseribe how a statistical model can be used to underpin a run off triangles epproach 5.2.7 Discuss the assumptions underlying the application of the methods in §2.1 {0 5.2.6 above. Page 2 @IFE: 2019 Examinations _ OVERVIEW ‘This booklet covers Syllabus objectives which relate to run-off tangles. Inthe chapter on run-off triangles, you need to be able to calculate the reserve required by a company for claims that wil arrive in the future. There ‘ae three methods required, namely the chain ladder method, average cost pper claim and Bomhuetter-Ferguson. You also have to be able to adjust all of these methods for inflation. ‘Questions on this chapter tend to be based on numerical calculations of the reserve, ‘The inflation question that crops up most often in the exam is the inflation adjusted chain ladder method. ‘Another popular question in tho exam is to ask about the assumptions Underlying each of the methods, so make eure that you learn your booiawerk. © IFE: 2019 Examinations Poge 3 CORE READING Allofthe Core Reading forthe topics covered in this booklet is contained in this section. ‘The text given in Arial Bole font is Core Reading. Chapter 24 — Run-off Triangles Run-off triangles (dolay triangles) usually arise in types of insurance (particularly nondlfe insurance) where it may take some time after a lose until the full extent of the claims which have to be paid are known. It is important that the claims are attributed to the year in which tho policy was written. ‘The insurance company needs to know how much itis fable to pay in claims so that it can calculate how much surplus it has made. However, it may be many years before it knows the exact claims total ‘There are many causes for the delays in the claim totals being finalised. The delay may occur before notification of the claim and/or between notification and final settlement. It is clear that although the insurance company does not know the exact figure for total claims each year, it must try to estimate that figure with as much confidence and accuracy as possible, ‘There are several ways of presenting claims data, which emphasise different aspects of the data. Here they will be presented as a triangle, which is the most commonly used method. The year in which the incident happened and the insurer was on risk is called the accident year. ‘The number of years until a payment is made is called the delay, or development period. Page + 2019 Examinations 7 The claims data Is divided up by the accident year and the dovelopment year. ‘The following table Is an example of claims data referenced by accident year and development year. In some types of Insurance it might be relevant to look at development of claims by month or quarter, but the principles are unchanged. Cumulative claim payments Accident Year Development Year 0 1 2 3 4 2008 786 1410-2216 2440 2,519 2009 904 1875-2518 2,798 2010 995 tea 2.880 2011 42200-2142 2012 4.482 Figure 1 8 The figures given are cumulative and represent total amounts paid by the end of each development year. They have been compiled after the end of the 2012 accident year. For the 2012 accident year, only payments with delay 0 have been reported. For the 2011 accident year, payments with delay 0 and delay 1 have been reported, and so on. The task is to decide the amounts yet to be paid in respect of the given accident years. This can be done for 2012 by looking at previous accident years. If the cumulative payments Increase In a similar way, It is possible to say that they are likely to be about 3,788 in 4 years’ time, This figure is obtained by assuming that the 2012 accident year is similar to the 2008 accident year in the pattern of making payments, and estimating cumulative payments at the end of development year 4 by: 2,519 1182%25 7 3,788 © FE: 2019 Examinations Pages ‘This is not necessarily the ‘best! estimate, but it is possible to fill in the lowor triangle in Figure 1 by comparing present figures with past ‘experience. This processis the main object of this chapter. 9 The method used above Is based on an accident-year basis where claims development is clustered by the year an accident has occurred. In that respect, such datawould include incurred but not yet reported (IBNR) claims. Another method of recording data is by underwriting year. This procedure would group claims by the timo policies were written rather ‘than when they occurred. A third method of grouping claims is by reporting year. 40 Consider a claim on an incident occurring on 1 December 2017 for an insurance policy written on 21" December 2016. Assume that this was reported to the insurer on 15th January 2018 and settled in July 2018. Thon, assuming calendar years are used, it would show: + indevelopment year‘ for 2017 under an accident-year basis ‘+ In development year? for 2016 under an underwriting-year basis + Indovolopment year for 2018 under a reporting-year basis. ‘The relative merits of each basis are discussed further in Subject SP7. 114 The basic assumption made in estimating outstanding claims concerns the run-off pattern. The simplest assumption is that payments will ‘emerge in a similar way in each accident year. 42 The proportionate increases in the known cumulative payments from cone development yaar to the next can then be used to calculate the expected cumulative payments for future development years. 13 Howaver, as the example below illustrates, there are a number of choices as to which such ratio should be used to project future claims. Pages © IFE: 2019 Examinations Note: The ratios that are used to project future claims are known as development factors or link ratios. Example Proportionate increases in cumulative payments Accident Development Year Year o 1 2 3 4 2008 786 1.794 1,410 1.572 2,216 1.101 2440 1.032 2,519 2009 904 1.742 1,575 1597 2515 1.112 2,796 2010 995 1.823 1,814 1.588 2.880 2011 4,220 1.756 2,142 2012 1,182 Figure 2 For each accident yoar from 2008 to 2011 there is a different ratio for the increase in cumulative payments from development year 0 to development year 1. It is not clear which is the ‘correct’ one to use ‘when projecting forward for accident year 2012. For a conservative estimate of cumulative payments, it might be best to take the largest ratio, fe 1.823. However, some sort of average of the ratios would seem more appropriate. It is possible to use a simple arithmetic average: 1,794 $1,742 41.823 41.756 a 1779 ‘The disadvantage of this is that it does not take into account that the years in which more claims occur provide more information. Thus, the ‘greater the amount of claims, the more confidence you can have in the ratio. ‘This suggests using a weighted average and the usual choice of weights are the cumulative claims values. © 1FE: 2019 Examinations Page? Accident a Ratlo Weight 2008 1.704 oe 2000 1742 as 2010 4.623 oat 2011 1756 7 4,794 x 786 £1,742 $04 + 4.823 995 44,756%4220 _ 4 77 786 + 904 +995 + 4,220 7 ‘This method of estimating the ratios which describe the run-off pattern is called the chaincadder method. The most efficient mode of calculating the ratios is given later. 14 The general form of a run-off triangle can be expressed as follows: Each entry, Cy, in the run-off triangle represents the incremental claim and can be expressed in general terms: Oy = Fy8Xig Hey whore: + fy 1s the development factor for year j , representing the proportion of claim payments by year j. Each ry is indopendent of the origin year F. + 5 is a paramater varying by origin year, , representing the ‘exposure, for example the number of claims incurred in the origin year i. + Xj.j 19 a parameter varying by calendar year, for example representing inflation, an error term, Pogo 8 ‘© 1FE: 2019 Examinations 15 The method of calculating the development ratios is demonstrated in the following example: Example Recall that the ratio in accident year 2008 was calculated as follows: 1410 4.794= Ee ‘The ratlos for the other accident years were calculated in a similar way. ‘The numerator of the equation at the end of paragraph 13 can therefore be written as: 1410 9g «4575: 994 4 1814, gg , 2142 1706 + 1575 904 4 4814 «995. 786 ia 904, 995 saa 4,220 1,410 $4,575 1,814.42, 142 44220 Thus, the development factor can be calculated using the cumulative claims in development years 0 and 4,410 41,575 44,814 42,142 "786 + 904+ 995 +1220 ‘The name given to this method presumably arises from the ladder-like ‘operations which are chained over the development years. The development factors for tho chain ladder technique can be found for ‘each development year by adding the appropriate number of terms. ‘This is Illustrated bolow. 6 Accident Tea Development Year 0 1 2 3 4 2008 786 140 2.216 2,480,519 2008 904 1,575 2,515 2,796 2010 995 4,814 «2,880 oii 1,220 2,142 2012 4,482 Figure 3 © IFE: 2019 Examinations Page o ‘Tho development factors (in order, starting with year 0 to 1) are: 6941 ag a77 BOIL ggg, 8298 4.407, 25" 1.092 3,905 nate a700 "A731 aa 2,440” 47 Development factors have been calculated for each development year. It is now possible to project forward each accident year. For accident year 2012, the projections of cumulative claims art 4ie2x4.77 100 4,9182x1.77 1,586 = 3,331 4.902% 4.777 1.586% 1-107 688 4,482 1.777 1.586» 1.107 «1.032 ~ 3,806 For accident year 2011, start from 2,142 in development year 1 and use ‘only the last 3 link ratios. Projections of cumulative payments: Accident Development Year Year ° 1 2 3 4 2008 2009 2,885 2010 3488 3,290 2011 3397 3,761. 3,884 2012 2aoo «3,331 3,688 3,806 Figure 4 48 Note that no projection can be done for the first accident year because ‘not possible to project beyond the highest development year. itis therefore assumed that all claims from this cohort have completely run off. Page 10 (© FE: 2019 Examinations. In practice, where this assumption may not be appropriate, manual adjustments in the form of ‘tail factors’ may be used. These are beyond the scope of Subject CM2, and will be discussed in Subject SP7. 49 The reserve that needs to be held at the end of 2012 Is the sum over all accident years for which a projection has been made of the difference between the cumulative payment at the end of development year 4 and the last known entry in the development triangle for that accident year. 20 So from Figures 1 and 4, the reserve at the end of 2012 Is: (2,885 -2,796) + (3,280 ~ 2,880) + (3,881-2,142) + (3,808~ 1,182 21 Note that no discount rate has been applied to the payments in different years. 22 The chain ladder technique Is used primarily to estimate the development of cumulative claim payments. However, itis useful to ‘check whether itfits reasonably with the claims data which have already been received. To illustrate this, look at the data in Figure 2. ‘To check how well the chain ladder technique performs, the claims in development year 0 for accident years 2008-2011 will be considered in the example below. Example ‘The actual claims In development year 0 are as follows: 2008 786 2009 904 2010 995 2011 1,220 ‘The development factors calculated in paragraph 16 were 1.777, 1.586, 4.407 and 4.032. Using these, estimates of cumulative claim payments in each development year can be obtained. Its of particular Intorost to ‘compare these with the actual values given in Figure 2. 9 FE: 2019 Examinations Page tt Hence, the following table gives the ‘fitted’ values using the chain laddar technique: Fitted cumulative claim payments Aceldent Development Yoar Year ° 1 2 3 4 2008 7e6 4,397 22152452 2,531 2009 904 1,606 2,548 2,820 2010 995 1,768 (2,804 2014 14220 2,168 Figure 5 23 It is now possible to compare Figure § with Figure 1. However, itis ‘preferable to look at the increases in cumulative payments when considering the fit of the modal. This gives a more sensitive test. ‘The increases in cumulative payments with development year (both actual and fitted) are given in Figure 6. Dovolopment Year 0 1 2 3 4 2008 © Actual «=«-786-=—«624.- 806. 22A 78 Fitted 786 «= 611, O18. 28779 Error I 9-12-13 2009 Actual «= «904 674940284 Fitted 904. «702, 942272 Error 7 1 2 9 2010 Actual «= «995 8191,066 Fitted 995773 (1,086 Error - 46 30 2011 Actual «= 1,220 922 Fitted 1,220 948. Error - -26 Figure 6 Page 12 @IFE: 2019 Examinations 24 None of the errors is large enough to suggest that the model is. Inaccurate. However, despite this check itis quite possible that the estimate ‘obtained may be a poor guide to the future. possible to adjust the calculated development factors in the light ‘of other Information. This method which uses prior knowledge can take a formal approach, but it is more often an ad hoc adjustment. 26 There may be good reasons to change the development factors. For ‘example, changes in accounting methods or claims administration can iter the speed with which claims are settled. This would give rise to ‘changes in the development factors and it would be sensible to reflect this in the estimate of future claim payments. 27 The development factors, either calculated directly from the data, or set using expert knowledge, are always used in the same way to estimate outstanding claim payments. 28 The chan ladder method can also be applied to a triangle of loss ratio Gata rather than cumulative payments, where the loss ratio for a given evelopment and accident year is the cumulative payment up to and including that development year divided by the total premium income in respect of the given accident year. 29 The chain ladder technique is based on the assumption that payments from each accident year will develop in the same way. In other word the same development factors are used to project outstanding claims for each accident year. 30 Changes in the rate at which claims emerge can only be incorporated by ‘hand adjustment’ of the development factors. © IRE: 2019 Examinations Page 13 31 The final assumption made when the chain laddor technique is used concerns inflation. Itis assumed that weighted average past inflation {will ba repsated In the future, This ia beoaues claims inflation is ane of the Influences swopt up within the projection factors. 132 Using the general statistical modal described in paragraph 14, it can be seen that the basic chain ladder takes the form: Oy =7)5) +04 ‘This might be an unrealistic assumption, and it will be considered in ‘greater detall In the following section. 38 When considering Inflation, itis important to bear in mind that itis claims inflation which is important. Thus, although a standard measure of overall inflation may be used, the inflation rate inheront in claims may be quite different. For example, a court decision can affect the size of claim payments. 34 Dealing with pastiinfiation Claims Inflation will affect the payments in the run-off triangle by calendar year of payment. 35 In the model considered hero, it will be assumed that claims inflation is ‘at the same annual rate for all claims within a particular calendar yoar of payment. 36 Each calendar year of payment corresponds to a diagonal in the triangle. For an illustration, look again at Figure 1. Page 14 © 1FE: 2019 Examinations 37 When adjusting for inflation, itis the payments in each calendar year which need to be considered, rather than cumulative totals. The first ‘stop is to calculate incremental payments from the cumulative totals, by differencing along each row. ‘The same operation was performed earlier and the following figure can be compared with Figure 6 in paragraph 23. 38 Example Figure 7 gives the incremental (or non-cumutative) claim payments for the data in Figure 1. Incremental claim payments in monetary amounts Accident mae Development Year ‘ 1 2 3 4 2008 7es 62h BUG 22K 2008 90d 67t_ 940284 2010 995 819 1,066, 2011 1,220 922 zz 1,182 Figure 7 39 Suppose that the annual claim payments inflation rates over the 12 ‘months up to the middle of the given year are as follows: 2009 5.1% 2010 6A% 2011 7.3% 2012 5.4% For simplicity, itis also assumed that payments are made in the middle of each calendar year. An index can now be calculated in order to convert all payments to mid-2012 prices. @ FE: 2019 Examinations Page 15 ‘The payments in Figure 7 can now be adjusted using the inflation rates. Figure 8 gives the inflation adjusted Incremental payment data, Incremental clam payments at mid-2012 pricos Accident Year 2008 2009 2010 2011 2012 ° 1,088 4.125 1,286 1.482 Development Year 1 2 3 4 T1912 8679 759 got 281 863 1,066 922 Figure 8 40 Now itis straightforward to form a table of inflation adjusted cumulative payments to which the chain ladder technique can be applied. ‘The forecasts of cumulative payments at mid-2012 prices are given in Figure 9. Forecasts of cumulative claim payments at mid-2012 prices Accident ‘Year 2009 2010 2011 2012 Page 16 Development Year 2 3 4 3,204 3341 3,432 3833701 (3,802 139-3434 (3,528 © IFE: 2019 Examinations 41 Dealing with future inflation The pradietions of cumulative payments do not, however, take account of future inflation, In order to forecast the actual payments, an assumed rate of future inflation will be needed. Again, itis necessary to convert to non-cumulative data rather than the cumulative totals before adjusting these for future inflation in a similar way to that used when dealing with past inflation. 42 Applying an annual inflation rate of 10% (at 30 June) to the data in Figure 9 gives revised forecasts of the cumulative claim payments as follows: Forecasts of cumulative claim payments in monetary amounts Accident Development Year Year 1 2 3 4 2009 2,090 2010 3196 3,306 2011 3435-3820 3,984 2012 2136 ©9455-3848 (3,986 Figure 10 43 The reserve that needs to be held at the end of 2012 is 6,136. 44 Assumptions underlying the method “The key assumption underlying this method is that, for each origin year, the amount of claims paid, in real terms, in each development ‘year is a constant proportion of the total claims, in real terms, from that ‘origin year. Explicit assumptions are made for both past and future claims, Inflation. @IFE: 2019 Examinations Page 17 45 Thorefore, using the general statistical model, the inflation adjusted chain ladder method takes the form: Cy = H8)X 165 +89 Average cost per claim method 46 This method, considers separately the two key elements of total claim amounts, ie the number of elaims and the average amounts of the claims. 47 This method roquires development table and claim numbers. for both total claim amounts 48 A third development table, of the average claim amounts, is then formed by dividing the figures in the corresponding cells of the first two tables. 49 The next stage is the projection of figures in the average claims and number of claims tables, using either grossing-up factors or davelopment factors. ‘50 Finally, the projected ultimate claims can be calculated by multiplying together for each accident year the projected figures for the average claim amounts and claim numbers. 51 A reserve ean then be calculated by subtracting all payments to date in respect of claims relating to the data in the table. ‘An example is given below to illustrate this process. ‘The average cost per claim method is not uniquely defined. It may therefore equally be applied on an accident yoar cohort to elther paid or incurred claims, or on a reporting year cohort. Page 18 @IFE: 2019 Examinations 52 Itis, however, important to ensure that the form of the data for the umber of claims corresponds to that of the total claim amounts (ie pald claims corresponds to the number of claims settled and incurred claims corresponds to the number reported). Of course, itis also Important only to apply the method to data for ‘which the development is considered to be stable and hence suitable for projection into the future. It is more likely that the reporting of claims will be more stable than the settlement, so the example given below relates to claims incurred. 153 Further, the method is not uniquely defined as regards the particular form of grossing-up factors or development factors. Howover, the grossing-up method is generally considered simpler and is used ‘simple average form in the example that follows. 54 Finally, the method described above ignored any adjustment for inflation. This can, however, bo done in exactly the same way as the adjustmont to the basic chain ladder method to form the inflation ‘adjusted chain ladder method (io if the data being used have been adjusted for inflation, it would simply require an index for future inflation to be applied to non-cumulative projected average claim ‘amounts before multiplying by the projected claim numbers). In practice an adjustment for inflation would normally be mad “The following example is done on the basis that the data have not, and need not, be adjusted for inflation. Cumulative incurred claims data, by years of accident and reporting development by ° 1 2 3 4 § Ul 4 2777 3,264 3,452 3,596 3,719 3,717 3,717 2 3.252 3,804 3,973 4,231 4,319 AY 3 3,725 4404 4,79 4,946 4 4,521 5422 6,676 5 5369 6,142 6 5818 @IFE: 2019 Examinations Page 19 Number of reported claims, by year of accident and repor development by 0 1 2 3 4 5 oul 1 414 460.482, 488 49249494 2 483 506 526535839 AY 3 494 548 S72 (582 4 530 588615 5 545 605 6 587 ‘55 Dividing each cell in the first table by the corresponding cell in the ‘second gives the accumulated average incurred cost per claim. ‘Average incurred cost per claim, by year of accident and reporting development by ° 1 2 3 4 8 UF 1 6708 7.096 7.162 7.365 7.559 7.524 7.524 2 7A79 7.518 7553 7.894 8.013 AY 3° 7.540 8.036 8.355 8.498 4 8530 9.221 9.229 5 9851 10.152 6 10.445 56 These tablos lead to the grossing-up factors and projected ultimate figures given in the following table (the projections are based on the, undertined, simple averages of the grossing-up factors). Page 20 (© FE: 2019 Examinations Average claim amounts: ° 1 2 3 4 5 utt 6708 7.096 7.162 7.365 7.550 7.524 7.524 892% 94.3% 95.2% 97.9% 100.5% 100.0% 2 TAT 7.518 7.553 7.894 8.013 7973 90.0% 94.3% 94.7% 98.0% 400.5% 3 7540 8.035 8.385 8.498 8.632 ay ° 874% 934% 96.8% 98.45% 4 5530 9.221 9.229 9.657 883% 95.5% 95.57% 5 2881 10.182 10.766 915% 94.3% 5 10.445 11.699 89.28% Claim numbers: by 0 1 2 3 4 5 ult 1414 460-482 48849294484 83.8% 97.8% 98.8% 99.6% 100.0% 2 453 526536539 5A4 837% 93.5% 97.2% 99.1% 99.6% 3 494 548872582 588 84.0% 932% 97.3% 98.95% AY 4 530 553615 632 83.9% 93.0% 97.37% 5 545 605 649 84.0% 93.2% 6 587 64, 83.88% 57 The total ultimate loss is therefore the sum of the following projected ‘amounts for each accident year: Projected AY Average cost per claim x Claim Numbers = Loss estimate @1FE; 2019 Examinations Page 21 1 7.524 496 3717 2 7973 541 4343 3 8.632 588 5,076 4 9.657 632 6,103 5 10.766 649 6,987 6 11.699 664 7,768 Total Projected Loss Estimate = 33,964 58 If the claims paid to date amounted to 20,334, the total reserve required would be 13,630. 59 As there is no unique way of defining the method, there is no unique ‘sot of assumptions. In particular the assumptions relating to inflation will depend on the data used. {In general terms, however, there are the assumptions that for each origin year, both the number and average amount of claims relating to ‘each development year ara constant proportions of the totals from that origin year. Finally, itis worth noting that for the assumptions to hold for this ‘method, it would be normal for them also to hold for a simpler method applying to total rather than average claim amounts, such as the chain ladder method. Loss ratios 60 The ratio of incurred claims to earned premiums over a defined poriod Is called the loss ratio. 61 Investigation of the loss ratios for each of soveral different origin years: would normally shew some consistency, provided that there have not been any distortions, and in particular no significant change in premium rates. “The expected loss ratios will also have formed part of the derivation of the premium basis. Page 22 © IFE: 2018 Examinations. 62 Itis therefore logical that a loss ratio based on tronds of past data, underwriters" views, or market data, could be used as a basis for an estimate of the eventual loss and hence the outstanding claims. tis, however, on Its own a very crude measure due to the fluctuations that are inherent in any claims experience 63 The Bornhuettor-Forguson method combines the estimated loss ratio with a projection method. 64 It therefore improves on the crude use of a loss ratio by taking account of the information provided by the latest development pattern of the claims, whilst the addition of the loss ratio to a projection method serves to add some stability against distortions in the development pattern. 65 The concepts behind the method are: ‘That whatever claims have already developed in relation to a given origin year, the future development pattern will follow that experienced for other origin years. ‘The past development for a given origin year does not necessarily provide a better clue to future claims than the more general loss ratio. 66 Inits simplest form the concopt leads to the following approach to calculations: 4. Determine the initial estimate of the total ultimate claims from each origin year using premiums and loss ratios. 2. Divide these estimates by projection factors (f}) determined, ino normal manner, from a claims development table. These are effectively estimates of the claims that should have developed to date. 3. Subtract these amounts from the corresponding total ultimate claims figures to give an estimate of the amount of claims that are yet to develop. (© IFE: 2019 Examinations Page 23 67 Clearly, the three stages could be combined and expressed as: Future claims development = ( Promium Estimated Loss Ratio {1 \ 68 As the final estimate of the ultimate loss is based on observed data and ‘an Initial estimate ignoring the observations, this method could be viewed as using a Bayesian approach. 69 In its original form, the Bornhuetter-Ferguson method was applied to the development of incurred claims. However, it could equally be applied to the development of paid claims, using either an accident year or policy year cohort. 70 Further, the original projection was done using a chain ladder approach, although alternative development factors or grossing-up factors (g ) could easlly be applied instead (io g would replace + In the above expression). 71 The original form also made no explicit adjustment for inflation, although the method could be adjusted in a similar way to the other methods. 72 The example below is based on the original form of the method, but examiners would expect students also to be able to apply the method ‘to paid claims. ‘The first stage is to determine the development factors, using the same mothod as far the chain ladder methods. Page 28 @IFE: 2019 Examinations Cumulative incurred claims data, by years of accident and reporting development by o 4 2 3 4 5 Ul 1 2866 3,934 3503 3,624 3.719 3,717 9,717 2 3359 33989 4093 4291 4,919 3 3948 4503 4779 4,946 4 4073 5422 5,676 AY 5 5369 6,142 6 5818 TOTAL 25,933 23,200 17,991 12,801 8,038 3,717 ToTAL- lastno: 20,115 17,148 12,315 7,855 3.719 RATIO. {r) 1458 1.049 1.039 1.023 0.999 1.000 DEV FACTOR 1.290 4.114 1.062 1.022 0.999 () 73 Next, the expected Ultimate Loss Ratio, say 63%, is applied to the ‘eamed premium (EP ) to give the initial estimate of the total ultimate oss (UL) for each accident year. Initia estimate of total ultimate losses, by accident ye ay 1 2 38 4 5 6 EP 4486 5,024 5,690 6,590 7,402 8,502 UL 14 6.210 (ossee) 9729 4470 4714 5470 6.210 7.057 74 (NB: In this example, the expected loss ratio has been taken as that ‘experienced for the fully developed first accident year. This has beon ‘done purely on the grounds of lack of other information.) (© IFE: 2019 Examinations Page 25 75 The next stage Is the application of the development factors to the estimated ultimate losses and the addition of the incurred claims that have already been reported. Revised estimate of total ultim: ay 6 4.290 0.225 7,057 Emerging liability 1,588 Roported liability 5,818 Ultimate lability 7,406 8 4414 0.102 6210 633 6,142 6775 4 4.062 0.058 5,470 317 5,676 5,993 3 1.022 0.022 ama 104 4,946 5,050 losses, by accident year 2 0.999 0.001 4,170 4 4319 4318 4.000 0 3,723 0 3,7 3,717 76 The total ultimate lability relating to these six accident years is, therofore, 33,256. If the claims paid to date amounted to 20,334 (the same figure that we used before}, the total reserve required would be 12,922. 77 Again the assumptions depand on whether the original or an amended version of the method is being used. For the original method, the underlying assumptions are the same as for the basic chain ladder method, together with the assumption that the estimated loss ratio is appropriate. Page 26 © FE: 2019 Examinations PAST EXAM QUESTIONS _ This section contains all ofthe past exam questions from 2008 to 2017 that ‘ave related to the topics covered in this booklet These questions are taken from Subjact CT6. Solutions are given later in this booklet, These give enough information for you to check your answer, including working, and also show you what an adequele examination answer should look like. Further information may be avaiiable in the Examiners’ Report, ASET or Course Notes. (ASET can be ‘ordered from Acted.) We first provide you with a cross reference grid that indicates the main subject areas of each exam question. You can use this, ifyou wish, to select the questions that relate just to those aspects of the topic that you may be particularly interested in reviewing. Alternatively you can choose to ignore the grid, and instead attempt each ‘quostion without having any clues as to its content. © FE: 2019 Examinations Page 27 peyduony Tay oF ABE ‘Buuedwog 1ePOn suondunssy wane aa D odov 2ePPET WIE ‘sio9eh dn Bulsso19 “Si0pe wawdoranog 5 6 Gross reference grid ss SAAB S pp! eie|sleizieleleleieiginisl © IFE: 2019 Examinations. Page 28 ‘Subject CT6 September 2007 Question 4 ‘The table below gives the cumulative incurred claims by year and eared premiums for a particular type of motor policy (Figures in £000s). Claims paid to date total £18,000,000, The ultimate loss ratio is expected to bo in line with the 2003 accident year. Accident Year 2003 2004 2005 2006 ° 3,340 3,670 3,690 4,180 Development year 1 2 3780 4,270 4080 4,590 4,290 7 Eamed Premiums 4.400 4,800 4,900 5,050 5,200 Ignoring infietion, use the Bornhuetter-Ferguson method to calculate the tolal reserve required fo meet the outstanding claims, assuming that the ‘claims are fully developed by the end of Development Year 3. 8] © 1FE: 2019 Examinations Page 28 ‘Subject CT6 April 2008 Question 5 “The following taste shows the claim payments for an insurance company in Units of £5,000: Accident Development year year 0 1 2 3 2004 410 84278 2005 575940281 2006 814 1068 2007 42 The inflation for a 12-month period to the middle of each year is given as follows: 2005 = 20062007 5% 55% 5.4% The future inflaton ror 2007 is estimated to be 6% per annum. Chains ate fully runoff a the end ofthe development yoar 3. Caloulato the amount of outstancing claims arising from accidents in year 2007, using the inflation adjusted chain lader metho fo Subject CT6 September 2008 Question 2 ‘Write down the general statistical mode! for the run-off tangle ciaim data ang explain the ters used, co) ‘Subject CTE September 2008 Question 5 “The table below shows the cumulative values (In units of £1,000) of incurred claims on a potflio of an insurance company: Undenenting Development year year 1 2 3 2008 3541 zai 9.801 7006 2949 6,850 2007 3.804 Page 30 © IFE: 2019 Examinations ‘The estimated loss ratio for 2006 and 2007 is 87% and the respective premium income (also in units of £1,000) is: Premium income 2008 14,044 2008 11314 2007 12,549 Given that the total of claims paid to dato is £20,103,000, calculate the reserve for this portfolio using the Bomhuetter-Ferguson method. a Subject CT6 April 2009 Question 6 “The following information is avallable for a motor insurance portfolio: ‘The number of claims settled: Development Year Accident year 0 1 2 2006 442 154 50 2007 623 1 2008 681 ‘The cost of settled claims during each year (in 00's): Development Year Accident year 0 1 2 2006) 6321 4,901 701 2007 7012 2237 2008 7278 Claims are fully run off after year 2. Calculate the outstanding claims: reserve using the average cost per claim method with grossing up factors. Inflation can be ignored, {19} (© IFE: 2019 Examinations Poge St ‘Subject CT6 September 2009 Question 8 ‘The cumulative inourred elaims for an Insurance company for the fast four accident years are given in the following tablo: Development year ‘Accident year ° 1 2 3 2005) 96 136 440 168, 2008 100 156 160 2007 120, 130 2008 136 1k can be assumed that claims are fully sun off after three years. The premiums received for each yoar from 2005 to 2008 are 175, 181, 190 and 196 respectively. Calculate the resarve at the end of year 2008 using: (@) The basic chain ladder method. {b) The Bornfuetter-Ferguson method, 112) ‘Subject CT6 April 2010 Question 8 The table below shows the incremental claims paid on a portfolio of Insurance policies together with an extract from an index of prices, Claims are fully paid by the end of development year 3. Development Year Price index Accident Yer 0 1 2 9 Year — (mid year) 2008 103 321382008 100 2007 2 16 2007 104 2008 11035 2008 409 2009 192 2008 111 Galoulate the reserve for unpaid claims using the inflation-adjusted chain ladder approach, assuming that future claims inflafon will be 3% pa. [14] Page 22 (© IFE: 2019 Examinations ‘Subject CTS Saptomber 2010 Question 9 ‘An actuarial student has been working on some claims projections but some fof her workings have been lost. The cumulative claim amounts and projected ullimate claims are given by the following table: Accident Development Year Utimate Year =O 1 2 3 1 40011485 1,762, x 2 4250 'Y 1,820 41,8623 3 13302 1,805 2122.5 4 Zz 22788 All claims are paid by the end of development year 3. It is known that ultimate claims for accident years 2 and 3 have been ‘estimated using the Basic Chain Ladder method, (). Caloulate the values of W, X and Y. fo) For accident year 4 the student has used the Bomhuetter-Ferguson method using an earned premium of 2,500 and an expected loss ratio of 90%. (i) Calculate the value of z. 4 (i) Calculate the outstanding claims reserve for ell accident years impliod by the completed table io Trotal 10] © IFE: 2019 Examinations Poge 33 ‘Subject CT6 April 2011 Question 11 “The toblo below chowe cumulative claims paid on a portfolio of insurance policies. Development Year ° 1 2 3 2007 240 eid 302 305 Acoldent — 2008 260 320 322 Yoar 2008 270 3128 2010 776 All claims are fully run off by the end of development year 3 (Calculate the total reserve for outstanding claims using the basic chain {adder technique. m ‘An actuary is considering modelling the future claims assuming that individual development factors are lognormally distributed with the following parameters: Development Year oto1 1102 2t03 # 0.171261 0.035850 9.008787 o 0.032148 0.045606 0.046853 (@) Show that under these assumptions the cumulative development factor {0 ultimate is also lognormally distributed. (il) Calculate a 99% upper confidence limit for the outstanding claims rolating to the 2010 accident year. fo Trotal 1] Pago st (© IFE: 2019 Examinations 10 Subject CT6 October 2011 Question 10 ‘The table below shows cumulative claims paid on @ portfolio of motor insurance policies. Developinent Year 2 Accident Year o 1 3 2007 120 134 146 148 2008, 140 780 185 2008 135 149 2010 138, All claims are fully run off by the end of Development Year 3, (Calculate the total reserve for outstanding claims using the basic chain ladder technique. i ‘An actuarial student suggests an alternative approach to projecting the claims as follows: ‘+ For each of Development Years 1 to 3 caloulate the observed development factor separately for each accident year. ‘+ Then project claims assuming the development factor for a given year is the maximum of the observed development factors for the relevant accident year. = For example for the development factor from Development Year 1 to Development Year 2 we can observe actual factors for Accident Years 2007 and 2008, To project claims, we assume that the development factor for Development Year 1 to Development Year 2 is the maximum of the two observed factors. (i) Colouiate the increase in the reserve for outstanding claims if claims are projected in this way. 8] (ii) Discuss why the method in (i) may not be appropriate, a Total 14), © FE: 2019 Exarinations Page 95 11 Subject CT6 April 2012 Question 8 Tho table belew shows claims pald on a portfolio of general insurance policies. You may assume that clalms are fuly run off after three years. Underwriting Year 2008 2009 2010 201 ° 450 503 en 555 Development Year 1 2 3 312 417 a 369 162 436 Past claims inflation hae been §% pa. Howaver, itis expected that future claims infa will be 10% pa. Use the inflation adjusted chain ladder method to calculate the outstanding claims on the portiolio, Page 36 10} @1FE: 2019 Examinations 2 2 ‘Subject CT6 Soptomber 2012 Question 7 The table below shows claims paid on a portfolio of general insurance policies. Claims from this portfolio are fully run off after 3 years. amano [Bata ae zoo [a oe Ls a __2011 414 (i) Estimate the outstanding claims using the basic chain ladder approach. La You are asked to investigate the fit of the model by applying the development factors from part () fo the claims paid in development year 0 ‘and then comparing the fitted claim payments to the actual payments, Construct a table showing the difference between the fitted payments and the actual payments in the table above, 8) (ii), Comment on the results of the analysis in part (i. Q [Total 12), ‘Subject CT6 April 2013 Question 5 ‘The following table shows incremental claims data from a portfolio of insurance polices for the accident years 2010, 2011 and 2012. Claims from this type of policy are fully run off after the end of development year two. Incremental Development Year Claims 0 4 2 2010-2828 1.484388 Accident Year = 2011. 1,749 4,188 2122117 Estimate the total claims outstanding using the basic chain ladder technique. m © FE: 2019 Examinations Pago 37 1“ 15 Subject CT6 September 2013 Question 6 ‘The fables bolow show cumulative data for the number of claims and the total claim amounts arising from a portfolio of insurance policies. Claim Numbers Total Clsim Amounts Development Year Development Year o 1 2 o 1 2 oro a7 182151 2010 43,200 87,430 126,310 don 17188 2017 68,900 125,200 2012 99 2or2 74.250 Cairns are fully un off ater two development years, Eslimate the outstanding claims using the average cost per claim method with grossing up factors (10) Subject C76 April 2014 Question 9 “The table below seis out incremental claims data for a portfolio of insurance policies Develooment year, Accident year 7 , e 2014 goa. «535142 2012 arte Bit 2013, 1912 Past and projected future inflation fe given by the following index (measured to the mid-point ofthe relevant year) Year — Index 2011 100 2012 107 2013110 2014113 2015117 Estimate the outstanding claims using the inflation-adjusted chain ladder ‘technique. 8 Page 38 © IFE: 2019 Examinations 16 Subject CT6 September 2014 Question 5 ‘The table below shows the incremental claims incurred for a certain portfolio of insurance policies. Development year Accident year 0 i 2 201 2,238 4,388 600 2012 3,380 1,808 2013 4,996 Cumulative numbers of claims ere shown in the following table: Development year Accident year ; : a 2019 140 203 224 2012 180 230 2013 266 () Caloulate the outstanding claim reserve for this portfolio using the average cost per claim method with grossing up factors, ml (i) Stale the ascumptions underlying the calculations in part (). 2) [Total 10}, @ FE: 2019 Examinations Page 99 "7 18 ‘Subject CT6 April 2015 Question 2 ‘The table below shows cumulative claim amounts Incurred on a portfolio of insurance policies, Accident Development Year Year o 1 2 3 2011 1,509 1,969 2,108 2.207 2012 4,542 2,186 2.985 2013, 4734 i924 2014 4973 ‘Annual premiums written in 2014 were 4,013 and the ultimate loss ratio has ‘been estimated as 93.5%, Claims can be assumed to be fully run off by the ‘end of development year 3. Estimate the total caime arising from policies written in 2014 only, using the ‘Bomhuetter-Ferguson method. wm ‘Subject CT6 October 2015 Question 8 ‘The run-off tangle below shows curnulatve claims incurred on a portfolio of ‘general insurance policies Development Year 1 Policy Year ° 2 3 201 1,528 2,034 2212 2310 2012 i812 2251 2.961 2013 11698 1851 2014 2.125 ‘Annual premiums written in 2014 were 4,023 and the ultimate foss ratio has been estimated as 91%, Claims paid to date for policy year 2014 are 672, Estimate the outstanding claims to be pald arising from policies written in 2014 only, using the Bomhuetler-Ferguson technique, stating any ‘assumptions that you make. 13) Page 40 © IFE: 2019 Examinations 19 20 Subject CT6 April 2016 Question § (Explain why insurance companies make use of run-off triangles. (2) (i) The run-off tangle below shows incremental claims incurred on a portfolio of general insurance policies. : Development Year Policy Year 0 1 2 3 2011 4957 9.440 93 512 2012 6089 = 5.278184 2013, 5823 4,799 2014 7,224 Calculate the outstanding claims reserve for this portfolio using tho basic chain ladder method. (Total 9} Subject CT6 September 2016 Question 8 ‘The table below shows incremental claim amounts paid on a portfolio of ‘general insurance policies, where claims are assumed to fully run off after three years, Underwriting Year Development Year o 1 2 3 2012 804 286 110 35 2013, 21 302 120 2014 685 340 2018 801 Past and projected future inflation is given by the following index (measured to the mid-point ofthe relevant year) 2012 100 2013 403 2014 105 2015 108 2016 105 2017 107 2018 410 Estimate the outstanding claims reserve using the inflation-adjusted chain ladder technique. (12) OIFE: 2019 Examinations Page 41 2 ‘Subject CT6 April 2017 Question & (Write down the genaral farm of a statistical model for a claims cun-off triangle, dofining all torms used. 5] “The table below shows the cumulative incurred claim amounts on a portfolio of insurance policies. Development Year] Tnaervming Year| 0 1 Z| 2014 g2i5 | esa7_| 10.078. 2015 2.886 | 7.123 2016 4467, Claims are assumed to be fully run off after Development Year 2. The: fstimated loss ratio of both 2015 and 2016 is 91% and the respective premium income in each year is: Premium Income 2014 11,965 2015 12,012 2016 12,887 The total of claim amounts paid to date is 21,486 from policies written in 2014 to 2016. (i) Calculate the outstanding claim reserve for this portfolio using the Bomnhuetter-Ferguson method io (Total 14), ‘Subject CT6 September 2017 Question § ‘The table below shows the cost of claims settied per calendar year for a set ‘of car insurance policies, wit figures in €000s. ‘Devalopmant Year ‘Accitont Year @ 1 2 2014 5819 ‘308 238 2016 3234 | 1,088 2016 7119 Page 42 (© 1FE: 2019 Examinations ‘The corresponding number of settiad claims is as follows: Development Year “Accident Year 0 1 2 2014 760 8 37 2015 819 98, 2016 381 {) Coloulete the outstanding claims reserve for this portfolio, using the m1 average cost per claim method with grossing up factors. (il) State four key assumptions made in part (0). © IFE: 2019 Examinations (2) {Total 9} Page 43 ‘SOLUTIONS TO PAST EXAM QUESTIONS ‘Subject CT8 September 2007 Question 4 foie 4409 timate L400 2 91.6667% The ultimate loss ratios 7555 “The development factors from the run-off tangle are: 400 + 4400 oa04 4,270 i. 4,270 4,590 2 $7TO%4.890 5 43455 and 3,750+4,080 ee 3,750+4,080-4,290 3750+ 4,080 44.200 9,340 +3,670+3,690 ids For Accident Year 2004, tho initial estimate of the ultimate claim amount ig 0.916667 x 4,900 = 4,491.67 . Using the development factor for Year 2 10 Year 3, the expected cumulative claims incurred at the end of Development Year 2's: 4491.67 ASSLT 4,358.98 1.03044 ‘This is less than the actual figure at the end of Development Yoar 2 by 4,590 4,358.96 =731.04 , so the adjusted expected ultimate claim is: 4,491.67 + 231.04 = 4,722.71 For Accident Year 2005, tho intiat estimate of the ulate claim amount {s 0.918667 x5,050= 4,629.17 . Using the development factors for Year 4 to Year 3, the expected cumulative ped claims at the end of Development Year 1 is: 4,628.17 1.03044 1.13155, er Page 4s © IFE: 2019 Examinations This is less than the actual figure at the end of Development Year 1 by 4,200 -3,970.14 = 319.86 , so the adjusted expected ultimate claim is: 4,629.17 + 319.86 = 4,949.02 For Accident Year 2006, the initial estimate of the ultimate claim is 0.916867 x8,200 = 4,766.67 . Using the development factors for Year 0 to Year 3, the expected cumulative paid ciaims at the end of Development 4,766.67. _ 766.87 __ 3609.10 1.03044 x 1.13455 1.132710 poe ‘This is less than the actual figure at the end of Development Year 0 by 4,150 ~3,609.10 = 540.90 , so the adjusted expected ultimate claim is 4,766.67 +540.90 307 56 ‘The Bomhuetter-Ferguson estimate of the outstanding claims reserve is the total of the expected uitimate claims incurred less the claims paid to date: 4,400 + 4,722.71 + 4,949.02 + 5,307.56 ~ 18,000 = 4,978.30 “Therefore the outstanding claims reserve is £4,979,000, ‘Subject CT6 April 2008 Question 5 We already have incremental data, so next we adjust for past inflation (io change the figures into present day mid 2007 values): Claim payments in Development year 2007 money terms Pr ti (€5000s) 6 7 7 $ Hoxie | aR toes 2004 | stances | otase” | 8895 | aa 5 470.703 | 905.144 bs g 5108 1 yo, gape © | zoos | ance | Rt | 201 : 639.383 : Bant088 < 2006 ‘857.956 1,086 zor [_ 11a ‘© IFE: 2019 Examinations Page 45 Accumulating the data gives: Claim payments in evelopment year 2007 money terms (€5000s) 0 1 2 3 2004 | 47a703 | 1,383.847 | 1,611.511 | 1.690.511 2005 | 639.383 | 1,620.43 | 1,911.143 2006 | 957.956 | 1,923.956 Accident year 2007 | 1,142 We calculate the development factors as folows: se0ostt | DY 293 08H 1.049022 gon. seteter.ta3 py ig telugteters 4 168768 919.947 1830.143+1973 958 DY 0-1 Straseeaesssiasrass "7498907 Using these to project the 2007 accident year values only gives: Claim payments in Development year 2007 money terms (€5000s) 0 1 2 3 2004 | 478.703 | 1,383.847 | 1,611.51 | 1.690.511 2005 | 630.383 | 1,630,148 | 1,911.143 2006 | 957.956 | 1,923.956 Accident year zoor | 1142 | 2,883,752 | 3,335.373 | 3,490.001 ‘The ineremental future claims for the 2007 accident year are: Claim payments Development year 2007 money terms — 5000) | 1 2 3 Ay | 2007 | aaiize2 | 4eie2t | 163.508 Page 46 © IFE: 2019 Examinations ‘Adjusting for future inflation (fe calculate the actual money 10 be paid in future years): ‘Claim payments in Development year 2007 money terms, (£50005) a : 7 = "iriicrea 108) 999.021 1.08 163.508>1.08° ay 07 20 1,848.69 | set76_| 205.97 ‘So the estimated total future amount for outstanding claims is +1848,69 + 561.76 +208.97 = 2,616.4 So the reserve is: 2,616.4 x £5,000 = £13, 082,000 (6 SF) Subject CTS Septomber 2008 Question 2 “The general staistical model is: Op = Kj +H where + jis the accident year + [ts the development year + Gj'sthe incremental claims amount «F's the development factor for year J representing the proportion of claims paid by development year/ «9,18 parameter representing te exposure (for example the numberof aims incurred) iy 18a parameter varying by calendar year (for example representing infiation) + 9 tsanerror term. @ FE: 2019 Examinations Page 47 ‘Subjact CT6 September 2008 Question 5 We need to caleulaie the develnnmant factors, Year 2 to Year 3: 9,01 74 336098 Year 1 to Year 2: 7.411+6,850 3541+ 2,948 = 2.181158 For accident year 2006: Expected ultimate loss is 0.87%11,314 = 9,843.18. Expected loss to date fs £843.18 4 1.336098, fee So the actual loss is fess than the expected loss by: 7,367.10-6,850 = 517.10 Finally, the adjusted ultimate toss is 9,843.18-517.10 = 9,326.08. For accident year 2007; Expected ultimate oss is 0.87% 12,549 = 10,917.63. Expected loss to date is ‘So the actual oss is greater than tho expected loss by 3,894-~3,798 55 ~ 05.45 Finally, the adjusted ultimate oss is 10,917.68 + 95.45 =11013.08 ‘This means that the reserve is: 9,501-+9,326.08 + 11,013.08 -20,103 77 , jo £9,737,000. Page 48 FE: 2019 Examinations, 5 ‘Subject CT6 April 2009 Question 6 First we need the cumulative figures for the claims settied each year: Cumulative claim amounts (£0008) Accident Development Year Year 1 2 0 2006 «6.321 «8.222 8,928 2007 «7012 «9,249 2008 7,278 Cumulative claim numbers Accident Development Year Year 0 4 2 2006 «442583 BAD 2007623784 2008 (88t rst table by the corresponding figures in ‘We now divide the figures in the ft ‘cost per elzim on a cumulative basis: the second to calculale the average Accident Development Year 1 Year 0 2008 14.30090 13.86809 13.8714 2007 © 14.25522 12.6082 2008 © 10.68722 Using grossing up factors fo estimate the utimate ACPC for each accident year: Average cost per claim (cumviative) Accident Development Year ‘Year a 1 2 20s 1030537% 99.0132% 100K 1430090 13.86509 19.67714 89.2436% | 99.9132% zocor “2een2 12,60082 261176 96.203% 2008 558722 14.11830 © FE: 2019 Examinations Page 49 ‘Since AY 2008 is fully run off, the final figure, 13.87744, is 100%. Hence, ‘wo can calculate what percentage each of the previous values are of ti final igure. For example, 13.2650 is {AEE = 99.9102% ‘The grossing up factor for DY 4 is 99.9192%. So the estimate of the is 2m ultimate figure for AY 2007 ts 42G0082 = 12.61176 [Now we have the fal figure for AY 2007, we can now find the percentages for the previous value, 1}2552% = 09.2436% ‘The grossing up factor for DY 0 i the simple average of tho 2006 and 2007 DY O figures 122052709283 - 96,1487% . So, for AY 2008, theultimate 1.11530 Using a similar approach for claim numbers: Claim numbers (cumulative) Accident Development Year Year o 7 68.7403% | 92.2240% 100% 2o0¢ 42 503 643 78.2773% | 92.2240% zoos 782TT 2240% 795.90870 173,5086% 200s 7850E 926.4200 ‘The total amount expected to be paid in cleims sottied is: £8,923 +(12.61176x795.8867)-+(11.11580% 928.4200) = 20,258.00 From the cumulative claim amount triangle we have already paid out a total of 8,923+.9,24947,278 = 25,450 ‘So the outstanding claims reserve is: 29,258.00 25,450 = 9,808 = £3,808, 000 Pago 60 © IFE: 2019 Examinations ‘Subject CT6 September 2009 Question 8 (Basic chain ladder method Development factors: Year 010 Year 1 1364186 +190 4 395443 96+ 100+120 1401160 Year 1 to Year 2: 1404160 7 ar 110 Year Muy = 02739 Year2 to Year 3: Ware a0 PEELEeeeCeeEeerCPrs eeeer ec ceeees eae See eeeeeeeerrzrrcerl Teckdert | Development year veo | O11 2 z 2005 | 96_| 98 [140 768 2006 | 100 | 156 | 160 60.2 19, 2007 | 120 | 130 “Taos 027997 x12 = 160.274 2008 | 136 TOEN1 SABAS x1, 027397 1.2 = 225.918 ‘The reserve is: 492-160-+ 160.274 ~ 130 +223,915 ~196 = 150-189 i) Bornhuetter-Ferguson method “The ultimate loss ratio is: For accident year 2008, the expected ultimate foss i: 0.96 181=173.76 Wiking backwards using the development factor for year 2 f0 Year 3, the Sypacted Ioss atthe end of development yoor 2s: 113.76 173.76 _ 444, iB T8 1448 ‘© IFE: 2019 Examinations Page 51 ‘The actual figure at the end of development year 2 is greater than this by 160-144.8 = 18.2, so the adjusted expected ultimate claim is: For accident year 2007, the expected ultimate loss is: 0.98% 190 = 182.4 Working backwards using the development factors for year 2 to year 3 and for year 1 to year 2, the expected loss at the end of development year 1 is: 182.4 1824 2 147,9467 1.2%1.027397 fee ‘The actual figure at the end of development year 1 is less than this by 147 9467 ~130 = 17.9497 , so the adjusted expected ultimate claim is: 182.4~17.946) 164.4533 For accident year 2008, the expacted ultimate loss is: 0.98196 = 188.16 Working backwards using all the development factors, the expected loss at the end of development year 0 is: 198.16 4.2.x 1,027397 x 1.33544; Siceant “The actual figure a: the end of development year 0 is grealer than this by 196—114.2832 = 21.7168 , so the adjusted expacted ultimate ciaim is: 188.16 + 21.7188 = 209.8768 So, the reserve usirg the Bornhuetter-Ferguson method is: 188.96 -160 + 164.4593 ~130+209.6768-106 = 197.28 Page 52 (FE: 2019 Examinations Subject CT6 April 2010 Question & [Adjusting the Incremental data for past inflation to obtain the figures in 2000 prices using the index: ee Bevalopment year, 3 7 Zz z soa ah xa zoos | ie | Me | Be 18 asaa3_| 94.154 | 20.532 Accident | 2007 or Be 16 year g3.e23_|_ 21.385 2008 38 __{_ 112.018 2008, 73. ‘The cumulative figures at 2009 prices are: Development year a 1 2 a BOs [4s [148.484 17B.076 “| 191.016 ‘Accident (72007 | 83.923 | 715.308 | 131.308 year 200d | 12018 | 147 018 2009, 732, We now calculate the development factors. They are: ap.as4s 115 308147018, ss asecameaeriizore C0270 ‘Year 0to Year 1: rear 2; TBOM1208 Year 1 to Year 2: HAQISSIS2S5S = 1.17261 saiate Year 210 Your 3: 2486 = 1.07308 FE: 2019 Examinations Pago 53 ‘The projected cumulative figures at 2008 prices are: Development year 1 o rT [2 | a 1 Accident 2607 740.897 | vyeor (2008 F7z355 | 84.984 [por ra 198 5128. 040) ‘By subtraction, the projected incremental figures at 2009 prices are: Development year a 1 2 Zz [2cee Accident [2007 5.58) year 2008 Baie _|~ 12.689 2608. “arate | 29.225 | 14.498 ‘The projected incremental figures adjusted for future inflation are: : Bevelopmant year a 7 2 oT [28 [ Tae 7 oe 9877 Oaed 2008 25:376~1.03 | 42 509 x 1.09"| ce 26.198 | 13.356 ae xt 08 7 =| 2009 ‘20228 1.03] 14.4993 1.08 30.435 |" 31.005 | 16.943 and the estimated reserve is the sum of these projected figures: 9.877 + 26,198 + 13.956 + 98.495 431.005 + 15.843 = 194.7, Page st (© IFE: 2019 Examinations 8 Subject CT6 September 2010 Question 9 (0 Calculate the values of W, X and Y Using the Basic Chain Leder method the development factors are: 449547 +1805 __ 3,200+¥ 1 Ada5¥ +1805 __ 3.20054 DYOI0 tT Torr 425041902 3688 4,762.44820 _ 3,582. ertee qaae+Y 1485+ we 0Y2 037765 from AY2 wo eee that the frat DYZ figure of 1,820 was projected to 4,862.3. Hence: w 1820 afeg” 18623 = W 303.0 cinco the uimate figure X fa simply the DYS figure for AY ite equal to w. Hence: x =1803.0 From AY3 we see that the final DY? figure of 1,805 was projected 1° 2,122.5. Hence: 2.582, We 22,1228 er 4,465 + “768 Substituting n the value of W, we get ¥ 316920 @ FE: 2019 Examinations Page 65 (i) Calculate the vatue of Z ‘The inital UL fe given by: 2,600x0.8 Using the values of W and ¥ from part (j), we get the cumulative development factor for DYO to be: 73200416920, 3.582 __, 1803.0 3563 1485 + 16820 ~ 4762 6290 Jeans \ enor tatty 2.2802(1- => ultimate Habilty=2+868.8=2,2788 => Z=14100 (il), Outstanding claims reserve reserve = (ulimate claims) — (claims paid to date) = (4808.0 + 1,962.9 +2,122.5+ 2,278.8) ~ (1808.0 + 1820+ ,805+1,410.0)= 1,226.6 ‘Subject CT6 April 2011 Question 11 (0 Basie chain ladder Development year Accident year 0 1 2 3 2007 240, 214 902 305, 2008 260 320 322 2009 270 3129 2010 276 Devfactors 1.1874 1.03758 1.00993 Cumulative dev factors 1.24426 1.04789 1.009034 Page $8 (OIFE: 2019 Examinations Hence, the ultimate claims for each accident year are: 2007 305 2008 322x1.00993, zone 312.9x1.04789 = 927.8835 2010 and the total reserve for outstanding claims is: 1949,41649 +927.88955 +325 18868 -276~912.9-322= 85.60 (i) Legnormal development factors We are given that the development factor from year i to year /-+1 is lognormally distibuted, fe d, ~ lognormal (4.0?) Hence the cumulative development factor fj for accident year J is the ‘product of lognormal development factors, So ia Gf] where d~gnomat nc?) Infy= = Did; a Smano?) A But tne eum of independent normal random variables ts also normal, with parameters 4’ = Sa, and o® =o? 6S roma WS | itt r so mt,-a | and f © IFE: 2019 Exeminations Page 57 (i), Confidence Fit We need te find the distribution of the 2910 cumulative development factor: 3.4, 0.171281-+0.035850 +0.008787 = 0.215088 (032148? +-0,045606? +0.04685%" ~ 0.072867 Le?= Hence, forg ~ lognormal (0.215886, 0.07286"). Let U be the upper limit forthe outstanding claims. We require: Pleutstanding claims < U) (ultimate claims ~claims paid < U) =| P (276 fox -276

    You might also like