In this paper we study the effect of weather conditions on electricity consumption in Perm region. We employ the data on consumption, air temperature, humidity and wind power for each 3 hours in 2011-2015. Using nonparametric two-step procedure, we revealed non-linear relation between electricity and temperature. We also revealed the different shape of relation for periods with switched on and switched off central heating.In this paper we study the effect of weather conditions on electricity consumption in Perm region. We employ the data on consumption, air temperature, humidity and wind power for each 3 hours in 2011-2015. Using nonparametric two-step procedure, we revealed the “heating effect” for temperature periods cooler than 17°C and “cooling effects” for periods warmer than 17°C. We also revealed the increase in consumption and lower sensitivity of consumption to temperature changes for period with switched on central heating.
This article is dedicated to multivariate comparison of big number of GARCH, ARFIMA and HAR-RV families’ models considering their one-day ahead realized volatility, which is known to be a consistent measure of daily volatility. A total of 102 models from three families were included in comparison.Comparison was completed with the help of Model Confidence Set test using 3 different loss functions on 10 Russian stock assets, including eight stock assets and two stock market indices. Received results strongly suggest HAR-RV superior performance to other two families of volatility models on Russian stock market and confirm local findings of previous studies.
The paper compares the nowcasting quality of a range of models of Russian GDP using high-frequency data. The models compared are MIDAS in different specifications, including models with regularization and dimensionality reduction using principal components and Mixed-Frequency Bayesian VAR with Minnesota prior. Indices corresponding with GDP by production components are used as explanatory variables. Nowcasts of MFBVAR models are shown to have higher accuracy then obtained by any type of MIDAS models on different test time periods. We also analyze dynamics of nowcasting errors of models and calculate monthly estimate of GDP growth rate that can be obtained with MFBVAR models
The paper aims at finding the most accurate VaR model for the four most liquid Russian stocks. Among the possible VaR modeling techniques, the estimates considered in this work are based on GARCH models with six different distributions. A back testing analysis is performed to evaluate the accuracy of the alternative models and to find the worst predictable period in terms of the volatility behavior.
To compare the quality of admissions to various educational programs in higher education institutions we propose to use the adjusted demand curves. These curves are derived from the results of unified state exams (USE) of tuition-free students and students enrolled as olympiad winners without entrance examinations. To eliminate the influence of the admission exam composition (USE subjects and number of exams) in various programs we propose two methods of correction exams scores. The first method assumes a linear correction of the total USE score of individual student. This method is based on the known mean values and standard deviations of scores for individual subjects. In the second method, a nonlinear correction of the USE score for each individual subject is produced. In this method the standard distribution is preselected and its distribution parameters are used when USE scores are recalculated. As an example, a beta distribution is used as a standard distribution.
Classification problems for univariate and multivariate observations are often encountered in statistics and economics. However, all existing approaches to solving these problems have several essential drawbacks:
1. All these methods cannot help in testing the null hypothesis of no different classes;
2. The number of classes is assumed to be known a priori;
3. Theoretical justification of performance effectiveness of these methods is lacking.
In this paper a new nonparametric method is proposed which can help us to solve these problems. This method enables us to construct consistent estimate of an unknown number of classes and to test the null hypothesis of no different classes. Besides theoretical findings, we present results of experimental analysis of this method including comparison of its characteristics with the maximum likelihood method and k-means method in different situations.
In the paper relationship of men’s income and their marital status is tested empirically. The Russia Longitudinal Monitoring Survey of HSE (RLMS HSE) data (1994–2011) confirm that men who are officially or civil married and were married before earn more in average than men who are not and were not married. This is explained by hypotheses of selection and treatment. The hypotheses are tested by the Mincerian type equations estimates. The using data do not reject the hypotheses.
In the paper we use the RLMS-HSE 1994–2016 data to estimate changes in life satisfaction caused by cohabitation and marriage. We made empirical estimates by discontinuity design within the classical linear regression, ordered choice model, and model based on dichotomisation of dependent variable. The positive effects of changes in marital status on women’s life satisfaction are observed in all the cases but the marriage after living together. For men, the only effect of starting cohabitation is robust to the model choice and reduction of the width of observations window.
The paper proposes a two-step methodology of investigation the impact of technical efficiency (estimated by stochastic frontier model) on Russian industrial companies’ risks of financial distress (estimated by bankruptcy prediction model — see King, Zeng, 2001). We show that growth of technical efficiency has robust, significant and large impact on the expected probability of financial distress. We also extend the «benchmark» specification of bankruptcy prediction model by including dummy variables for structural breaks in bankruptcy dynamics associated with significant changes in Bankruptcy law.
In the paper technical efficiency of Russian plastic and rubber production firms in 2006–2010 is estimated by SFA. It is demonstrated that increasing firm size will cause increase in its efficiency and also there is increasing returns to scale in the sector. This result is robust for various specifications of the production function and SFA models. Autocorrelation of efficiency estimates is considered as a measure of their persistency.
We consider the structure of social network of university students, and analyze factors that lead to the personal communication (friendship) network formation. Using the data from student survey and administrative information, we estimate the econometric models that assume homophily and propinquity effects on the likelihood of being friends. Specifically, we estimate linear and logistic probability models with and without fixed effects. The results confirm the significance of such friendship formation factors as sharing the same study group, living in dormitory, similar academic achievement and having the same gender. The probability of a tie increases with the number of similar attributes.
In this study, we analyze correlations between the co-authorship network parameters and citation characteristics in Google Scholar. We estimate the count data regression model in a sample of more than 30 thousand authors with the first citation after 2007. There is a positive relationship between scholar’s citation counts and number of co-authors, between citations and the author’s centrality, and between scholar’s citations and the average citation of co-authors. The h-index and i10 index are significantly associated with the number of co-authors and average citation of co-authors.
What effect does change of the marital status of the individual to his wages in Russia? We used matching estimators (simple matching estimator and propensity score matching) on RLMS data for 2000–2009. On the basis of estimates we analyzed different transitions of changing marital status («not married – registered marriage», «registered marriage – divorced», «divorced – civil marriage», «divorced – registered marriage», «registered marriage – widow (widower)», «registered marriage – civil marriage») and their effect on wages separately for men and women. It was found that women are more influenced by cast of change in private life on the behavior on the labor market. For women, most of the transitions lead to some costs in wages, except a «divorced – registered marriage». For men it was significantly affect wages only transition «registered marriage – divorce».
In this work the demand for the incoming tourism in the Russian Federation is modeling. The panel data for 16 countries - the basic sources of tourist streams - and the period with 2000 for 2009 are used. Modeling is spent separately for each of 10 tourist zones of Russia. In quality a determinant of demand there are considered a total national product in a country of origin, the exchange rate, transport charges, cost of residing, lag of the demand variable and the fictitious variables reflecting influence of shocks in quality a determinants of demand. The received estimations of dynamic models of demand correspond to expectations, are statistically significant and can be useful in practice of planning of development of entrance tourism in various municipal formations and regions of Russia.
Analysis of factors, characterizing globalization, on transformation of higher education system, called as the «fundamental shift» by Kwiek (2001), is conducted with the use of data for 195 countries for the period 1988–2016. Models of dynamics of share of youth, getting higher education, before and after positive demand shock in higher education are considered. Globalization of markets and technologies development intensify competition stimulating growth of demand on highly-qualified workforce and increase of selectivity of universities in order to provide high quality higher education.