Sl. No, 0907130 HES/Isscsiee419916 STATISTICS Paper—IV Tins Ate Toss Hows] [Maximum Maro 28] INSTRUCTIONS Please read each of the following instructions carefully before attempting questions. ‘There are FOURTEEN questions divided ‘under SEVEN Sections. Candidate has to choose any TWO Section: attempt all the questions therein. All the Sections carry equal marks. -- ‘The number of marks carried by a question/part is indicated against it. Unless otherwise mentioned, symbols and notations ‘have their usual standard meanings. Assume suitable data, if necessary and indicate the same clearly. All parts and sub-parts of a question are to be attempted together in the answer book. Attempts of questions shall be counted in sequential order. Unless struck off, attempt of a question shall ‘be counted even if attempted partly. Any page or portion of the page left blank in the answer book must be clearly struck off. Answers must be written in ENGLISH only. Normal Distribution, ¢, F, Gamma and cv of Weibull ‘Tables are attached with the question paper. A graph paper sheet is attached to this question paper which may be carefully detached for answering graph-related question and then must be securely attached with the answer book. nd, 188 (PTO.1. “Thea (VSection—A. ( Operations Research and Reliability ) (a) Solve the following two-person zero-sum game using graphical procedure 1-3 305 “1 6 41 2 2 -5 0 (6) If the demand for a certain product has a rectangular distribution between 4000 and 5000, then find the optimal order quantity if the storage cost is @1 per unit and the shortage cost is @7 per unit. () A supermarket has two salesgirls at the counter. If the service time for each customer is exponential with mean 4 minutes and if people arrive in a Poisson fashion at 10 per hour, then find the expected percentage of idle time for each salesgirl. (@) Find out ML estimators “of the parameters in the density f pxtte P, x>0,b c>0 based on the following censored scheme : Xap Xap Xian where n and r are respectively the sample size and number of censored observations. A-GSE-P-TUD/88 2 10 10 10 102 @) (a) (b) Defining a location-scale family, outline the procedure for obtaining the coefficient of BLUE’s. 10 A computer contains 10000 resistors. When any resistor fails, it is replaced. The cost of replacing a resistor individually is ®1 only. If all the resistors are replaced al the samme Une, the cost per resistor comes down to 35 paise. The percent of surviving, S(t), at the end of f-th month is ri @ aa ¢ ss S(t): 100 97 90 70 32 15 0 Derive the optimum replacement plan. 25 A random sample of 25 observations drawn from Weibull model (with shape parameter = 2 and scale parameter = 4) is given. Obtain the moment estimates of the parameters : 25 18487 03761 07500 3053013545 1802 1570017708 + «1.3892 3.0466 17961 15319-05903 0628806461 16560 -7172-«193101.0509-1.6173 13162 0-705 1888918889 41505 A GSE P TUD/se 3 [PT.0.3. 4. Section—B { Demography and Vital Statistics ) (a) Discuss the sources of demographic data in India and also point out the uses and limitations of these data. fo) Explain the stationary and stable population models. Discuss the situation when stationary and stable populations are identical. () Explain briefly the uses of life table. In usual notation, prove that (d) Define crude death rate, specific death rate and standardized death rate. Interpret these rates. (e) State the general procedure and steps for the construction of life tables. (a) Explain an abridged life table and discuss its different columns. Discuss the King’s method for its construction. (b) What do you mean by fertility? Define crude birthrate, general fertility rate, specific fertility rate and age-specific fertility rate. How are these rates computed in practice? A-GSE-P-TUD/s8 4 10 10 10 10 10 25 255, Section—C { Survival Analysis and Clinical Trials } fa) Define hazard function and survival function. Obtain the same for an expo- nential distribution. (b) If f(q, F(t) and A(t} are the density, distribution and hazard functions of a random variable T, then show that Fj=1-F) Also establish a suitable relationship between hit) and reliability function. () If p, is the probability that a person aged x will survive to age x+1, then show that , p, +,@, =1, where been aPg dy (@) Discuss some specific situations in which it would be difficult or inefficient to perform clinical trials. fe) What are the principles for ethical clinical trials? 10 10 10 10 10 A-GSE-P-TUD/S8 5S [PT@) (e) () @ Define Type-] and Type-I censoring schemes. Obtain the maximum. likelihood estimator of parameter 0 in exponential distribution under the above censoring schemes. Also obtain the Fisher information for the parameter. Discuss various phases involved in a clinical trial. Also discuss the pros and cons of each phase. Section—D ( Quality Control ) Distinguish between process control and product control. Discuss the situations where they are used. Discuss various sources of assignable causes and --random causes of variations. Also state how they are detected in a manufacturing process. Define the terms ASN, AOQ, ATI and ARL for an acceptance sampling plan. What is an acceptance sampling plan? Discuss a single sampling plan (7, o, where the sampling is carried out using a binomial model. Find P,, if n=10, c=3 and p=0-05. A-GSE-P-TUD/58 6 28 a 10 10 10 10fe) (b) () (a State the importance of exponentially weighted moving average charts. How are these used in practical situations? 10 Discuss the importance of control charts for variables. 5 Obtain the control limits for (X, R) and (&, $) charts when standards are known and unknown. 20 In usual notation, show that 2 _ Isa ~ =f aa n |in-/2 where C, is a constant used for constructing control limits. 5 Discuss in detail the double sampling plan (N, nC, M9, ¢,) stating the assumptions followed in both the stages. Hence or otherwise, obtain the ASN function of this sampling procedure. 20 Section—E ( Multivariate Analysis ) Define p-variate normal distribution and obtain its characteristic function. Assume mean vector 1 and dispersion matrix 5. 10 A-GSE-P-TUD/S8 7 [ P70.0) () (a fe) 10. (a) (e) Define Hotelling’s T? and mention its application. Show that T(p, m)=— PP, im pet 10 Explain. the importance of principal components and discuss the method for extraction of principal components. 10 Define Wishart distribution and obtain its characteristic function, 10 Define canonical variates and canonical correlation. Give your interpretation. . 10 State the chief properties of Wishart distribution. If A - Wp ny), i then show that has the Wishart distribution with parameters ¥ and n, where k n=n, a assuming the independence of A;’s. 25 Define’ multiple correlation coefficient and obtain its non-null distribution. 25 A-GSE-P-TUD/s8. 81. Section—F { Design and Analysis of Experiments } (a) What are the basic principles of design of experiments? Explain them and their roles. (2) Define completely randomized design, and mention its merits and demerits. Also give its statistical analysis. () Define Latin square design with its applications, advantages and disad- vantages, (d) Explain analysis of covariance with its applications. Give an illustration for the use of analysis of covariance in identifying the response variable (y) and concomitant variable (x) fe). Bxplain split-plot design and give some situations where split-plot design can be suitably adopted. (a) What are factorial experiments? The following table gives the layout and the results of 2°-factorial design laid out in four replicates. The purpose of the experiment is to determine the effect of 10 10 10 10 10 A-GSE-P-TUD/S8, 9 [P-ro.different kinds of fertilizer, Nitrogen (N), Potash (K) and Phosphate (P) on potato crop yield Block 1 Block 2 nk kp mp|[kp pk nk 291 301 312 373||407 324 272 306 1 & nkp|| n nko mp1 101 265 106 450|| a9 449 338 106 Block 9 Block 4 pl mp kp |[np nk np 323_87_324 423||361 272 103 324 nk kn nkp|| kl nkp kp 334_279 128 471||302 131 437 435 Analyze the design and = draw conclusions. 25 (6) What is confounding? A 23-factorial experiment is conducted in 2 blocks of size 4 each, in 3 replicates. The ~arrangements are as follows-: Replicate | Replicate 2 | Replicate 3 Key Biock (1), ¢ ab, abelt), a be, abel(i), b, aq, abe| Identify the confounded effect in each replicate. 10 () Define a Randomized Block Design (RBD) specifying the statistical model associated with the design. Carry out a complete analysis of RBD with one missing observation. 15 A-GSE-P-TUD/88 10(e) « @ fe) Section—G ( Computing with C and R ) Given a five-digit positive integer, write a C program to find the sum of individual digits. (For example, if the given number is 96785, the required sum is 94+6+74+8+5=35) 10 Write a C program to pick up the largest tender from a set of tenders assuming that Tender-id and Tender-value for each tender are specified. 10 Express (A-B)*C in infix and postfix forms. 10 Write a program in R to plot Q-@ plot assuming that the two samples (Xj, 121, 2, +, 25 and ¥;, 1=1, 2, 250) are drawn from normal population, 10 Given n positive integers, write R code to find their ranks. 10 A-GSE-P-TUD/58 ry [P-T.0.14. (a) Mlustrate call by value and call by reference, and advantages as well as disadvantages of each method. Write a C function to evaluate the series without using a built-in function : 25 (2) Given X~ Niu, 6), where p and o? are specified, write a C program to compute PoC>q=[ Density of Moar A-GSE-P-TUD/88 12 SB6—7300evee auneo] ano7130 Normal Distribution Table aie Gat Gee Gens aN aoe Str si ee sn eis 600 eoes 79 7054 Tome 7123 rae ‘Sor an asa Toor Toe Teta me Hao en toe fem roe ree re = jer 0 tooo feat fone dest at a a se oi Se we i Ege ee Sa Se 8 oe Ee a Be ae $8 Sow Sees fees fess Ge eet ar Stat Soar fase se Gest Sos tary ume hos te fee a B57 gm mH sD oH Bee oe Se 25 20 SE SB Se BS Se ee Sei; Sees Sess soot Ser sere ome uma aoeo ree Sh Sia foe gat a me eo ore Set Sr are a77a s7ea 97 s709 a750 som ote ania sot? wr Se Gee See Ss Le Se tee Sk Se SS Si See ern es te fer soe oor form Sess fee oc) sume eset ato ae fein imeo pact eco eet Sots eat Srna Sobe no eee ee eee geez ge foot soe Sos ero crt fora sora arn See Soe eve art wr re iro ore Soot oat Soot ose Soot Sma Sone Sree ed neo Sos oem ee ee ee Sn Sor Sonn fet Soret eee ee Soe ae Sse oma boos sont Some Soot uct owe G08 os st So ess two ee ot ees oe Soe aor Soar _Sosr_sest_Swt_Sant_toot_teor_ ery Soose P-7up Table mont] 30 fxs a0 tas too tas tors tow to0s Coop Fanos 950 025 020 045 040 005 0025 001 0.005 0001 onons socal 1.00 030 040030 0.20 040 0.05 0.02 0.01 0.002 0.001 fiom tats ios 135 tage San ann ues uae Sauer Sam ‘occ ayer sm tise tam ans Zon a3es omg sc un ae os it os i Ge ae see 2} sam oreo ure eco tare tsi 220294 gen tea gaat ‘Shep Rem okt tome fe iret bis ayia ran 7] thm fim am Sten 13 ine aie ite toe ix} Som ate Sas ssn iam ti flee Bast Srey haat ee —B) tar assess to 1b te SB es By tm Si Sie ie tit oe ler Fm nme toast ons? 1036 12m 1 1m50 206 2576 amen am 80% — a0 70H 0m op _waas WaT ‘Confidence Levelie am 1375 yeas 1136 08 es os sar S86 rr as Be ei0 2 798 tee 78 788 750 7 moe 800 nar 1089 rr 256 ea ere os ord 623 or S89 a5 87 Soe Sat re 5a Bas 599 538 a rees 978 245 338 600 ess a2 895 57% 888 ea 823 508 501 aoe ‘er ee on ae ee e0 a3 pert ‘Table of F Distribution an 1598 1139 ais 795 tot eae S00 Ser 521 Soe an 45 a ia ie an 05 doe 30 ea i882 os? 275 fab 608 See S22 86 prs aa ar toe 3a Soe 200 an 278 an 37 asso? se2n-s3 5907.10 2a 3029 10 1621 ra87 738 a3 500 338 307 fe aa 40 ie 38 38 376 an 350 358 a8 aa 323 an 2738 roNe sor oe fe 20 382 36 300 330 aa 322 318 38 203 Ba 1005 78 ser BGGESES B5S8 BREESE al‘Table of F Distribution Critical values oF P forthe 0-05 significance level‘Table for Gamma Function Tinj=ferertd, 1sns2‘Shape Parameter e Corresponding to the Population ‘Coefficient of Variation cv of Weibull Distribution was a3] 0s Tao | 0am 30 ‘cme ons, | ane tas | tame ts ‘soso ozo | ommmotgo | 08K 20 tows 0s | tame tas | osm 2s sam oo | emt | aate7 290 semi aas, |e tes | 030030 bo cay) namss ro) Onmh as neon cas | omtsr tas | tas 310 tat ow | osm tno} east us 1.9650 05S 0.5608 185 | (0.3479 3.20 1.7581 0.60 0.5474, 190 | 0.3430 3.25 woe ow | ene ae | om 30 lato | osm 200 | eae 35, tam | ost 20s | oan 0 toes tao | games ato) oss 8, ins oss | oa ais | tate 350 time 0 | oan aan | aes 358 lus Sas | oars nas | sis 3.0 oo tm | aden a00 | ants as cost tos, | ngms as | camer 0 oo tao | aaus 2a | amos cons tas, | mn oats | oat aD ows tas | ome 288 | tat 390 ont 130) aan | on as overs | oat nes | oaks 400Sagan SLNo. 9907130 ‘ot rite Rol Hoon tis sheet. EEGEEEE oe Fee a