Порівняння методів

	Метод Кростона для переривчастого попиту	Регресія звичайно найменших квадратів (ЗНК)	Пуассонівська та від’ємна біноміальна регресія
Галузь	Економетрика	Економетрика	Економетрика
Родина	Regression model	Regression model	Regression model
Рік появи	1972	2019	1998
Автор методу	J. D. Croston (1972)	Wooldridge (textbook treatment); classical least squares	Cameron & Trivedi (textbook treatment); Hilbe (negative binomial)
Тип	Intermittent demand time-series forecasting	Linear regression	Generalized linear model for count data
Основоположне джерело	Croston, J. D. (1972). Forecasting and Stock Control for Intermittent Demands. Operational Research Quarterly, 23(3), 289-303.	Wooldridge, J. M. (2019). Introductory Econometrics: A Modern Approach (7th ed.). Cengage Learning. ISBN: 978-1337558860	Cameron, A. C. & Trivedi, P. K. (1998). Regression Analysis of Count Data. Cambridge University Press.
Інші назви	Croston method, intermittent demand forecasting, Croston Yöntemi — Aralıklı Talep Tahmini	ordinary least squares, classical linear regression, linear regression, en küçük kareler regresyonu	count regression, log-linear count model, negative binomial regression, Poisson / Negatif Binom Regresyon
Пов'язані	4	5	4
Підсумок	Croston's method, introduced by J. D. Croston in 1972, is a time-series forecasting technique built for intermittent demand series in which periods of zero demand are frequent. Instead of forecasting the raw series, it models the size of demand when it occurs and the interval between demand occurrences as two separate processes.	Ordinary Least Squares is the classical linear regression method that explains a continuous outcome as a linear combination of predictors. It estimates the coefficients by minimising the sum of squared residuals, and under the Gauss-Markov assumptions these estimates are the best linear unbiased estimator (BLUE).	Poisson regression is a generalized linear model for count outcomes — events tallied as non-negative integers such as hospital admissions, accidents, or article counts. It models the log of the expected count as a linear function of the predictors, and is developed in the standard count-data treatment of Cameron and Trivedi (1998); when the counts are over-dispersed, the closely related negative binomial model (Hilbe, 2011) is preferred.