DOI: http://doi.org/10.15350/2409-7616.2020.4.15

Alexey V. Burkov –Doctor of Economics, Professor of the Department of Applied statistics and informatics, Mari State University, Yoshkar-Ola, Russian Federation, ORCID: https://orcid.org/0000-0003-3188-2862, E-mail: alexey.burkov@gmail.com

Ruslan V. Pshenichnov –Graduate student, Mari State University, Yoshkar-Ola, Russian Federation, E-mail: psheni4nov@yandex.ru

Abstract. The article discusses the application of the multiple regression method for modeling the conversion of mortgage applications in typological groups. The definition of typological groups is based on different phases of mortgage market activity and different conditions for granting loans. The open information of Sberbank was used as an information base for the study. The presented methodology will make it possible to assess the number of accepted mortgage applications depending on various parameters of the mortgage lending market. As factors influencing the number of accepted mortgage applications, the following characteristics of the market were considered: the number of accepted applications, the dollar rate, the weighted average interest rate and the average price per square meter. The paper considers several regression models at different stages of approval of applications and with different market activity. In this paper, the following types of multivariate linear models are considered: the direct selection model of significant factors and the model using the ridge regression method. Further, the analysis of the quality of the obtained models was carried out, and the most acceptable models were selected for each of the groups. Durbin-Watson coefficients, residual distribution histograms, scatter diagrams and average approximation errors were used as criteria for assessing the quality of the obtained models. To compare the quality of the constructed models, comparative diagrams of the coefficients of determination and the average approximation error were also built. In the conclusion of the work, the analysis methods used were compared with the methodology presented in the article and its effectiveness was proved. Since the presented methodology will make it possible to predict the load on mortgage lending managers and optimize the planning of mortgage activities.

Keywords: mortgage lending market, regression analysis, cluster analysis, linear regression, ridge regression, mortgage market activity, suspensive condition stage, conversion of mortgage applications, Sberbank.

References:

1. Kokin A.S., Oskolkov I.M., Trofimova D.S., Sitnikov R.R. Analysis of the current state of the mortgage lending market in Russia. Economy: yesterday, today, tomorrow, 2018, no. 5A (8), pp. 127-142. (In Russian) URL: https://elibrary.ru/item.asp?id=37056592
2. Savosina D.A. Development of the mortgage lending market in Russia. Ufa, Aeterna Publ., 2020 pp. 100-103. (In Russian) URL: https://elibrary.ru/item.asp?id=41880513
3. Sinichenko O.A. Features of preferential mortgage lending in Russia. Vector of Economics, 2020, no. 9 (51), pp. 22. (In Russian) URL: https://elibrary.ru/item.asp?id=44044878
4. Shumeiko A.A. Statistical analysis of the development of mortgage lending in Russia.  Moscow, Buki-Vedi Publ., 2017, pp. 68-73. (In Russian) URL: https://elibrary.ru/item.asp?id=29420114
5. Egorova A.A., Akimova E.M. Analysis of the situation on the mortgage lending market in Russia. Economics and Entrepreneurship, 2018, no. 12 (101), pp. 1087-1091. (In Russian) URL: https://elibrary.ru/item.asp?id=36722289
6. Domashchenko D.V., Shchegolev A.S. Banking indicators – leading indicators of socio-economic development of regions (on the example of the Central Federal District).  Federalism, 2019, no. 2 (94), pp. 138-155. (In Russian) URL: https://elibrary.ru/item.asp?id=38576129
7. Wang Z., Crook J., Andreeva G. Reducing estimation risk using a bayesian posterior distribution approach: Application to stress testing mortgage loan default. European Journal of Operational Research, 2020, no. 287 (2), pp. 725-738. DOI: https://doi.org/10.1016/j.ejor.2020.04.020
8. Sirignano J., Giesecke K. Risk analysis for large pools of loans // Management Science, 2019, no. 65 (1), pp. 107-121. DOI: https://doi.org/10.1287/mnsc.2017.2947
9. Aydamirova L.D. Features and problems of mortgage lending in Russia. Bulletin of Science and Practice, 2020, no. 6, pp. 413-419. (In Russian) DOI: https://doi.org/10.33619/2414-2948/54/55
10. Kartvelishvili V.M., Nikolaeva A.V. The market for securitization of mortgage lending. Bulletin of the Russian University of Economics. G.V. Plekhanov, 2016, no. 2 (86), pp. 122-129. (In Russian) URL: https://elibrary.ru/item.asp?id=25896417
11. Faizullin I.E., Salakhova A.R. Mortgage lending during the crisis. Russian Journal of Entrepreneurship, 2009, no. 5-1, pp. 99-102. (In Russian) URL: https://elibrary.ru/item.asp?id=29736472
12. Orlova I. V., Turundaevsky V. B. Short-term forecasting of mortgage lending. Statistics and Economics, 2013, no. 6, pp. 175-177. (In Russian) URL: https://elibrary.ru/item.asp?id=20902438
13. Orlova I.V. Econometric modeling of the consequences of the economic crisis of 2008-2009 on the rate of GDP growth in different countries. Fundamental research, 2020, no. 4, pp.  62-69. (In Russian) URL: https://elibrary.ru/item.asp?id=42831246
14. Babeshko L.O., Orlova I.V. Toolkit for modern econometric modeling. Moscow, Tsentrkatalog Publ., 2020, 336 p. (In Russian) URL: https://elibrary.ru/item.asp?id=42923596
15. Babeshko L.O. Econometric forecasting based on heterogeneous information. Moscow, Vega-Info Publ., 2016, 232 p. (In Russian) URL: https://elibrary.ru/item.asp?id=26880196

For citation:

Burkov A.V., Pshenichnov R.V. Modeling the conversion of mortgage applications by the method of multiple regression in typological groups. CITISE, 2020, no. 4, pp.162-177. DOI: http://doi.org/10.15350/2409-7616.2020.4.15