1

Software for Data Analysis and its Applications for

the Distance Courses of PROMETHEUS



N.M.Tovmachenko, I.M.Sichkar, O.V.Lukovich, O.O.Kurbatov

State Academy of Statistics, Accounting and Audit
of State Statistics Committee of Ukraine,
Kyiv National University of Taras Shevchenko

prometey@dasoa.edu.ua
http://www.dasoa.edu.ua/

2
PLANNING REPORT
Software for data analysis were successfully using in
education and different spheres for data analysis (surveys
and investigations; complex samples, marketing; research
activities in social sciences, economics, medicine; life-
insurance, databases and direct marketing; public
management). Software of the statistical package for social
siences (SPSS) and its application for data analysis is
considered .
It is described the application of the learning management
system PROMETEUS in State Academy of Statistics,
Accounting and Audit of State Statistics Committee of
Ukraine.
At Academy educational courses are developed in the
format of distance learning in accordance with the industry
standard of higher education of Ukraine.
3
PSP SPSS is appointed in the first turn for statistics-
professionals, because has enough vehicle of statistical
analysis, wholly compared for capacity from SAS. SPSS has
the developed input language, wide possibilities in the area of
management and presentation of information that are
complemented a flexible interface from SUBD and specialized
subsystem for support of the plotted output. SPSS is the
acknowledged software for educational establishments of
European community, and what is why acquisition of license
its use is also actual in connection with tacking of Ukraine to
Bolony agreement. Academic license to the use of package
SPSS allows Academy to introduce the modern methods of
information and communication technologies in an educational
process. SPSS is used for elaboration of DC from the applied
and social statistics, analysis of data, econometrics, sample
researches, economics, actuarial mathematics, creation of the
script-programs .
4

SPSS Complex Samples

The practical use of design of selective researches for the receipt of
estimations of mean and general value of descriptions of general aggregate
requires development of mathematical methods of construction of the proper
optimum plans. Therefore, except for the use of the standard software SPSS
authors of this article is created proper software that includes script-
programs of construction of optimum plans and theirs statistical analysis.
The followings methods are realized: 1) method of the proportional stratified
questioning in the conditions of homogeneity of variances with the expenses
of selective inspections for stratums; 2) method of the stratified questioning
with the calculation of size of stratum after the criterion of Neyman with
heterogeneous variances and expenses of selective inspections for stratum; 3)
method of the stratified questioning with the calculation of optimum size of
stratum with heterogeneous variances and different expenses of selective
inspections for stratum. In Tab. 1. calculations are resulted with the use of
the script-programs with the use of experimental information .
5
Tab. 1. Results of calculations of point and interval values of descriptions of
general aggregate [4]


Analysis of results Tab.1 shows exactness of evaluation after selective research
depends on the presence of a priori information. If a priori estimations of
variances and measuring expenses are known on every stratum, optimum design
are built after the method of the stratified for complex samples with the
calculation of optimum size of stratum in the conditions of heterogeneity of
variances and different expenses of selective researches for stratums.

6
Methodica of forming of selective aggregate of reporting units is developed with using
these algorithms for the inspection of small enterprises. Executions were formed after
basic and additional the indexes of stratifications: basic type of economic activity after
KVED ( classifier of types of economic activity, resulted on a picture ), mean of amount
of regular workers of registration composition, belonging, to object of KOATUU (
classifier of objects administratively territorial mode of Ukraine ) : 24 areas. 2 cities,
ARK: Autonomous Republic of Crimea, Vinnitsya region, Volyn region, Dnipropetrovsk
region, Donetsk region, Zhytomyr region, Zakarpattya region, Zaporizhzhya region,
Ivano-Frankivsk region, Kyiv region, Kirovograd region, Luhansk region, Lviv region,
Mykolayiv region, Odessa region, Poltava region, Rivne region, Sumy region, Ternopil
region, Kharkiv region, Kherson region, Khmelnitskiy region, Cherkasy region,
Chernivtsi region, Chernihiv region, Kyiv city, Sevastopol city.


7
The application of statistical methods for construction of forecast model for behavior of electorate
with use the package statistical programms SPSS on base data the Election of public deputy of
Ukraine 2006 year are described.
On being in 30.03.2006 (Central election committee-CVK) gave information of results of voting for to
Elections of folk deputies of Ukraine on March, 26, 2006. In particular, after electoral districts and
regions of (Classifier of objects administratively territorial mode (regions) of Ukraine- KOATUU)
have next statistical information in table.

8
In tabl. reg_ok- sign after which association of groups was conducted after objects
of KOATUU (Classifier objects Administratively Territorial Mode of Ukraine) and
proper electoral districts, blok_p sign after which conducted pogrupuvannya
results of affirmative vote by parties and blocks (examined : 1- party of regions, 2
block (BYT)-Block of Yuliya Timoshenko, 3- block Our Ukraine.
9
Estimation of parameters of econometrics model with using software
SPSS, in particular 3-th step method of least squares (MLS)

For parameters estimation of econometrics model for national economy of
opened type with government control [6] are used software SPSS, in particular 3-th
step method of least squares (MLS). Econometrics model for description of
development of national economy of the opened type with government control
(MEM-2) is built Lyudeke [6]. It is widely use in teaching, as realistically enough
describes the certain aspects of development of economy. MEM-2 is economic
model, that consists of four equalizations: three equalizations of conduct
(consumption, function of investments and function of import functions) and
balance equalization. 3-th step MLS estimation of parameters of econometrics
model consists in the account of possible cross-correlation of regressions them
residuales of different equalizations of the system at a statistical evaluation
simultaneously of all of equalizations of model of 3-th step MLS appears more
effective, than 2-step MLS, if casual residuales of different equalizations of the
system are mutually correlated, that them a covariance matrix is not diagonal.
Although it should be noted that and in this case the 2-step MLS estimates of
structural parameters of the system remain suitable. Two first steps of 2-step MLS
are saved in 3-th step MLS [7]. However got as a result of these steps, separately
for every, estimations of structural parameters are not final, but transferred on a 3-th
step. As a result of application of 2-step MLS [7] to the model of Lyudeke the
resulted forecast value are received in Tab.2., and for 3-th step MLS results of
calculations are resulted in Tab.3.

10
MEM-2 is economic model, that consists of four equalizations:
three equalizations of conduct (consumption,
function of investments and function of import functions)
and balance equalization.
Interdependent variables (Ct is the personal consumption; It is private clean
investments and fixed assets (without reserve of investments); Imt is an import; Yt is a
national income) and Ct-1 the personal consumption is for previous period; Ut-1
profits of population from entrepreneurial activity, dividends and income to taxation
for previous period; Imt-1 import for previous period; Gt government spending
plus public clean investments and fixed assets plus changes in commodity supplies plus
an export minus taxes plus subsidies ).
11
System of equalizations, which expresses endogenous variables
through exogenous named the resulted model. The resulted form for a
structural model is written down as follows:

+
+
+

+
+
+
+
+
+
=
+
+
+

+
+
+
+
+
+
+ +
=
+
+
+

+
+
+
+
+
+
+ +
=
+
+
+

+
+
+
+
+
+
+ + +
=





. G
c b a 1
1
Im
c b a 1
c
U
c b a 1
b
C
c b a 1
a
c b a 1
c b a
Y
; G
c b a 1
c
Im
c b a 1
c b c a c
U
c b a 1
c b
C
c b a 1
c a
c b a 1
c b c b c a c a c
Im

; G
c b a 1
b
Im
c b a 1
c b
U
c b a 1
c b b a b
C
c b a 1
b a
c b a 1
c b c b b a b a b
I

; G
c b a 1
a
Im
c b a 1
c a
U
c b a 1
b a
C
c b a 1
c a b a a
c b a 1
c a b a c a b a a
C

t
1 1 1
1 t
1 1 1
2
1 t
1 1 1
2
1 t
1 1 1
2
1 1 1
0 0 0
t
t
1 1 1
1
1 t
1 1 1
2 1 2 1 2
1 t
1 1 1
1 2
1 t
1 1 1
1 2
1 1 1
0 1 1 0 0 1 1 0 0
t
t
1 1 1
1
1 t
1 1 1
2 1
1 t
1 1 1
1 2 2 1 2
1 t
1 1 1
1 2
1 1 1
0 1 1 0 0 1 1 0 0
t
t
1 1 1
1
1 t
1 1 1
2 1
1 t
1 1 1
2 1
1 t
1 1 1
2 2 1 2 2
1 1 1
0 1 0 0 1 0 1 0 0
t
12
3-step a least-squares method (3MNK) estimation of parameters of
ekonometrics model consists in the account of possible mutual
correlations of regressions them tailings of different equalizations of
the system at a statistical evaluation simultaneously of all
equalizations of model:

where B- (mm) a matrix of coefficients |ij is at endogenous
variables, - (m) matrix of coefficients of ij at exogenous
variables, At(ot(1), ot(2), . . . , ot(m))T vektor of casual tailings
which do not correlate with exogenous variables. 3MNK appears
more effective, than 2MNK, if casual tailings of different
equalizations of the system (1) are mutually correlated, that them
matrix of covariaces is not diagonal. Although it
should be noted that and in this case 2MNK estimation of structural
parameters of the system remain suitable. Two first steps of 2MNK
are kept in 3MNK. However got as a result of these steps, separately
for every equalization, estimations of structural parameters are not
final, but transferred on a 3 step.
n t CX BY
t t t
, , 2 , 1 , = A = +
( ) ( )
m l i
l
t
i
t , , 2 , 1 , = A
E = o o
13
14
2. It is described the application of the learning management system
PROMETEUS in State Academy of Statistics, Accounting and Audit of State
Statistics Committee of Ukraine.

SDL PROMETHEUS is used in many HEE both abroad (Moscow
Economic Statistical Institute, Federal Service of State Statistics of Russia)
and on Ukraine (in the National Academy of Government Tax Service of
Ukraine and other HEE).
After an acquaintance with experience of leading HEE of Ukraine a choice
was done in behalf of SDL PROMETHEUS a version 4.2 [1].
SDL PROMETHEUS allows to create the multimedia distance courses
(DC), which can spread and on compact-disk, provides possibilities of
distance learning and testing of listeners and also has necessary facilities
for the management of educational establishment virtual activity (it is
important to underline that every version of SDL PROMETHEUS passes
localization Ukrainian).
15
3. At Academy educational courses are developed in the format of distance
learning in accordance with the industry standard of higher education of Ukraine.

Conception of the distance learning of State Academy of Statistics, Accounting and
Audit of State Statistics Committee of Ukraine is based on implementation of
normative documents of Department of Education and Science of Ukraine (the
Order of minister for education and science of Ukraine 802 from 04.12.2003
About claim of Measures on realization of Program of the distance learning
system on 2004-2006 years, by approval by Order of Department of Education
and Science of Ukraine of Statute About the distance education 40 from
21.01.2004 and others like that).
Academy is base higher educational establishment (HEE), which prepares
specialists for statistical services of Ukraine, has the system of the separated
subsections (branches, colleges, educational-consultative centers) ramified on
Ukraine, and also carries out scientific researches. Necessity of development of new
technologies of education at the ramified network and difficult educational-
scientific structure needed simple in administration, but the educational platform of
the system of the distance learning directed on the stream of many thousands of
students.
In addition, the system distance learning (SDL) must not require in relation to
technical vehicle requirements and infrastructure of decorated a pattern.

16
On implementation of orders Department of Education and Science of
Ukraine in relation to legalization of software and fight against illegal use, the
Academy has the licensed SDL PROMETHEUS [Fig.1]

17
PSP SPSS is appointed in the first turn for statistics-professionals, because has enough
vehicle of statistical analysis, wholly compared for capacity from SAS. SPSS has the
developed input language, wide possibilities in the area of management and presentation of
information that are complemented a flexible interface from SUBD and specialized subsystem
for support of the plotted output. SPSS is the acknowledged software for educational
establishments of European community, and what is why acquisition of license its use is also
actual in connection with tacking of Ukraine to Bolony agreement. Academic license to the
use of package SPSS allows Academy to introduce the modern methods of information and
communication technologies in an educational process. SPSS is used for elaboration of DC
from the applied and social statistics, analysis of data, econometrics, sample researches,
economics, actuarial mathematics, creation of the script-programs.
On implementation of orders Department of Education and Science of Ukraine in relation to
legalization of software and fight against illegal use, theAcademy has the licensed the
package of statistical programs (PSP) SPSS, that is actual in connection with tacking of
Ukraine to Bolony process. There are next components of software product for SDL:
methods of development of authoring educational tools; control the system by studies -
Learning Management System (LMS); the system of exchange information between the
participants of educational process; delivery of educational content system (as a rule web
site). In the use of standards of exchange educational materials there are follofing advantages:
expansion of market of services and acceleration of development; decline of prime price of
ownership
18
Cost of academic license for the enough complete set of the system SPSS
makes $ 3000.
19
DC of actuarial mathematics includes the followings modules: measuring of death rate
(functions of survival, force of death rate, tables of death rate); insurance annuities;
life-insurance; clean bonuses; backlogs of clean bonus; status of dozhitiya of all
persons from a group; status of dozhitiya of the last person is in a group; theory of
multiple dekrementa. Windows of distance course of life-insurance is on next fig.
20
Distance Course Social statistics consist of the followings 18 modules:
Object, task and organization of social statistics , International,
national organizations of statistical services, information generator, social
indexes, classifications and groupments, are in social statistics ,
Population as object of social statistics. Statistics of quantity, will make and
placing of population , Structure of families and Housekeeping,
Standard of living as object of statistical supervision , Statistics of profits
of population , Statistics of consumption the population of commodities
and services , Statistics of housings terms and domestic consumer
services of population , Statistics of social time of population, its
structure , Statistics of employment, unemployment and social defence
, Sanitary statistics. Statistics of health protection population ,
Statistics of settling apart of population, state and guard of environment and
rational use of natural resources , Statistics of level of formation of
population and development of the system of studies , Statistics of
organizations of science and scientific service , Statistics of culture, art
and rest , Moral statistics , political and public life , Methods of
mathematical statistics and their use are for the computer processing of data
of social statistics .
21
Fig.3. Windows of distance course of information systems and technologies in
statistics.
22
ANOVA (One way variance analysis)
23
The practical use of design of selective researches for the receipt of estimations of
mean and general value of descriptions of general aggregate requires
development of mathematical methods of construction of the proper optimum
plans. Therefore, except for the use of the standard software SPSS authors of
this article is created proper software that includes script-programs of
construction of optimum plans and theirs statistical analysis. The followings
methods are realized (Script- programms): 1) method of the proportional
stratified questioning in the conditions of homogeneity of variances with the
expenses of selective inspections for stratums; 2) method of the stratified
questioning with the calculation of size of stratum after the criterion of Neyman
with heterogeneous variances and expenses of selective inspections for stratum;
3) method of the stratified questioning with the calculation of optimum size of
stratum with heterogeneous variances and different expenses of selective
inspections for stratum. Calculations are resultes with using of the script-
programs on base of experimental information. DC Statistical methods of
samples research consists of the following modules: basic determinations of
methods of samples research; process of forming of samples; construction of
plan of selection; collection of information is for forming of selection; connection
of answers after a selection and forming of selective weights; frequencies and
descriptive statistics; statistical analysis of cross tables; construction of
estimations of relations; construction of estimations of parameters and them
confidence intervals for a general aggregate.

24
25
Table. List of educational courses of SPSS (prices are
indicated in u.e. without VAT)
26
In the environment of package for statistical analysis of data of SPSS students
on the examples of the real researches meet with the management of data,
calculation and interpretation of descriptive statistics and contingency table,
generalization of primitive experimental information by application of them
statistical treatment, by the construction of variance, regression, discriminant,
factor models, by the statistical analysis of time series of different type,
formulation and verification of statistical hypotheses, by the methods of
graphic and table presentation of results an analysis, at preparation of reports
and presentations, and practical lessons enable to get skills of work with
package SPSS. In particular, DC Statistical analysis and forecast of time
series consists of the followings modules: smoothingof time series; models of
ARPSS; seasonal models; models of multivariate regression; models of the
exponential smoothing; transmission functions of frequency and descriptive
statistics; application of models of time series; forecast of time series.

SPSS provides work on all of the stages of analytical process and enables a
researcher to establish intercommunications, receive results for acceptance of
optimum and quality-controlled decisions.

DC from the applied statistics exhaust after a method teaching on the daily
form of studies and done for the lead through of employments with the
Internet/Intranet.


27
RESUME
Academy base department HEE, that carries out preparation,
retraining and in-plant training, after specialities the applied
statistics, economic cybernetics, accounting and audit,
finances and credit, and prepares specialists for
establishments and organizations of the Statutory broker of
statistics, Ministry of finance, Government tax service, State
treasury, Government control-revision service, auditing firms,
state participating and commercial banks and other subjects of
activity. Experience of the use of facilities of the distance
learning is in the conditions of transition of HEE to the terms,
which answer the requirement of Bolony process proves that
the construction of DC allows to unify material which is
expounds out and to provide theirs listener without a direct
contact, and also use the format of the controlled from distance
learning for the in-plant of shots of enterprises training without
tearing away from a production.

DC are worked off after the method of teaching on the extra-
mural form of studies and redone for the lead through of
employments with the Internet/Intranet.
28
http://www.prometeus.ru
http://www.spss.com
http://www.sas.com
. . . 1, , 1996, 2. , 1999;
, . SPSS: .
. .-., 2002;
., .., .
. :
, , . II -
. 2004 .
.., .. i script-
SPSS . International conference modern stochastics: Theory and
Applications. - Kyiv, 2006 .
..
. : , ,
. III -
. , 2006 .
Garkusha V.Z., Tovmachenko N.M., Kurbatov O.O. The Application of the Learning Management
System (LMS) PROMETEUS in the State Academy of Statistics, Accounting and Audit (SASAA) of
the State Statistics Committee of Ukraine. : ,
, . III -
. , 2007 .
29

Dear colleagues!
Invite you to take part in the VI International
scientific and practical conference "System
of State Statistics in Ukraine: modern state,
problems, prospects", which will take place
on December, 5 in 2008 in the building of the
State Academy of Statistics, Accounting and
Audit State Committee of Statistics of
Ukraine.
Invitation http://www.dasoa.edu.ua