نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشجوی دکتری گروه اقتصاد نفت و گاز، واحد مرودشت، دانشگاه آزاد اسلامی، مرودشت، ایران
2 دانشیار گروه اقتصاد، واحد مرودشت، دانشگاه آزاد اسلامی، مرودشت، ایران
3 استادیار گروه اقتصاد، دانشکده اقتصاد و مدیریت، واحد شیراز، دانشگاه آزاد اسلامی، شیراز، ایران، پژوهشگر مرکز تحقیقاتی بتا، دانشگاه استراسبورگ، فرانسه
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
One of the problems of managing the program and budget of the country's municipalities is the lack of scientific forecasts of annual revenues and as a result of the structural budget deficit, and this issue has caused inaccuracies and reduced welfare and services to citizens, and finally the administration of It has caused problems and has led to the failure to achieve legal policies and objectives.Today, most municipalities in the country use simple scientific revenue forecasting methods or have developed simple experimental methods that ignore the impact of macroeconomic variables. According to the analysis of past conditions and the study of existing trends affecting municipal revenues, financial resources should be predicted and direct and indirect factors affecting municipal revenues should be considered.The purpose of this study is to predict the revenue of municipalities and Shiraz city on a case by case basis. The two methods used in this study are regression method by predicting macro-micro interface variables of VECM model and neural network model in macro-micro hybrid space using data from 2009 to the end of 1397. The present research is an applied research in terms of purpose and descriptive-analytical in terms of methodology and the method of implementation and collection of data is documentary and library.To compare the predictability of multiple regression model with artificial neural networks, two evaluation criteria of competing models including MSE and RMSE as well as independent t-test were used to differentiate the prediction values made by the model and the actual (observed) values and the results showed that the network Nervous has been significantly more accurate in predicting municipal revenue than the regression model.
کلیدواژهها [English]