Comparison of Machine Learning Algorithms in Demand Prediction Problem
This study is intended the major issues of applying econometric and machine learning techniques to a daily demand prediction problem. The purpose of the paper is going to be achieved via the models’ predictive power comparison on bakery retail chain data.
Подобные документы
- 1. Modelling of special warning indicators for banking crises: the evidence from developing economies
General comparison of signaling and econometric approaches, arguments for and against each of methods to predict a banking collapse. Econometric approach of banking crises' modelling. Fitting the prediction model for banking crisis 2008-2009 in Russia.
дипломная работа, добавлен 28.08.2016 Carrying out of econometric analysis by means of panel data models of the volume of innovative goods, works and services in the Russian Federation. The dependence on the number of personnel involved in scientific research and the internal expenses.
статья, добавлен 30.07.2017The problem of studying the relationship of economic indicators in economic analysis. Construction of an econometric model of the ratio of the exchange rate of tenge to the ruble and the possibility of using it to forecast real economic processes.
статья, добавлен 23.06.2020The predicting unobserved station-level demand for bicycles using both data mining techniques and stochastic modelling. Development of public bike-sharing systems, stochastic modelling of bicycle-sharing systems. Realization of peak detection heuristic.
курсовая работа, добавлен 28.08.2018The econometric models of goods export from Ukraine and the import to Ukraine in general, their seasonal and trend components are considered. The forecast of investigated indices for III and IV quarters of 2013 is made based on these constructed models.
статья, добавлен 28.09.2016Research a number of econometric models of the relationships between regional macroeconomic indicators of the labour market of Ukrainian economy. Evaluation of the developed models, the degree of influence factors, statistical analysis of the results.
статья, добавлен 14.08.2016Discuss a number of procedures that are designed for testing the specification of econometric models, concerned with the numerical properties of these OLS estimates and refer to certain properties of estimates as "numerical" if they have nothing to do.
монография, добавлен 08.03.2014The mathematical model of free market of three goods in the conditions of supply lag in the interval of discrete time with market parameters, demand vector and initial conditions is considered. The dynamics of allowable profit of the seller is obtained.
статья, добавлен 07.08.2022The formation of the decision support system in the field of regional development management have been considered. Different approaches in the framework of integration into the structure of adaptive simulation models of problem-oriented knowledge bases.
статья, добавлен 26.08.2021Pre-crisis, post-crisis windows definition. Bipower variation and jumps. Models for volatility forecasting. Forecast comparison. Value at risk estimation. Statistics on volatility estimators. Regression estimations. Forecast results of standalone models.
курсовая работа, добавлен 11.02.2017Electronic virtual markets can serve as an alternative tool for collecting information spread among numerous experts. Two motivation approaches compared via participant's activity represented by both the volume of trades and the number of transactions.
статья, добавлен 28.09.2016Main trends in the development of local markets for crop products. Analysis of methodical approaches to the application of econometric methods to determine promising directions of development of these markets. Systematization of the factor indicators.
статья, добавлен 28.09.2016European options on several assets. The Black-Scholes-Merton model. Numerical method. Sparse grid combination technique. Applying the method to the option pricing problem. Solving the transformed equation. Monte Carlo simulations. Bermudan styles.
курсовая работа, добавлен 26.08.2016Cyclically adjusted price-earnings ratio - one of the most valuable insights for a better measure of publicly held company's long-term financial performance. The forecasting model which based on the method to test predictors of the equity premium.
дипломная работа, добавлен 01.12.2019The analysis of modern approaches to forecasting of innovative development. The building of the predictive models in economic systems. The use of methodology of overcoming of temporal gap for calculating the extremum points of cycles of reproduction.
статья, добавлен 26.08.2016The significance of these results is the possibility of increasing the accuracy of the forecast of exchange rates on the Internet market by using advanced mathematical apparatus, based on the set of factors affecting the fluctuations of exchange rates.
статья, добавлен 30.07.2016Discusses the possibility of modelling individual factors determining inflation, using intelligent technology based on decision tree and matrix convolution. Constructed of the simplified model of the "manufacturers demand to profitability level".
статья, добавлен 25.09.2016Policy countries in a situation of migration of skilled labour and opportunities for social optimum achievement. Development of a model showing the interaction between the countries in the game, where public policies impact on the socio-economic balance.
доклад, добавлен 30.08.2016A focused factory is a approach for manufacturing enterprise. The development of heuristic scheduling algorithm in step-by-step method. Nagare cell is a combination of manual and semiautomatic machine for maximum output flexibility for productions.
статья, добавлен 17.04.2013Ideas of vanguard investigators about applying probability models and applications (stochastic processes and inference, distributions and characterizations, Bayesian inference, selection methods, regression methods) in health and economic research.
книга, добавлен 25.11.2013In this study problem of short-term forecasting for coal and crude oil production in Ukraine. Autoregressive and optimal filtering algorithm for linear systems based upon autoregressive model of second order were constructed for short term forecasting.
статья, добавлен 29.07.2016Study of algorithms of processing of nonlinear signal. The definition of the Gaussian model of normal random variables. The distribution of order statistics. The absolute value of the deviation of the weighted median. A balanced set of calculations.
книга, добавлен 25.11.2013Analyze the efficiency and the overall performance of Turkish airports using the data envelopment analysis. Efficiency scores under constant returns to scale. Indentifying the most and least efficient airports in Turkey using data envelopment analysis.
статья, добавлен 27.09.2016The most effective model to evaluate the enterprise innovative capacity subject to an uncertainty factor is the model based on the fuzzy sets theory. The model has obvious advantages in comparison with the expert and statistical methods of evaluation.
статья, добавлен 28.09.2016The likelihood ratio test is a test used to compare the fit of two models, one of which is a special case of the other. The test is based on the likelihood ratio, which expresses how many times more likely the data are under one model than the other.
статья, добавлен 10.09.2012