﻿ r 计量经济分析代写 Term Project Assignment 预测模型分析 - R语言代写, 数据分析, 经济作业, 统计作业

# r 计量经济分析代写 Term Project Assignment 预测模型分析

2019-09-04 11:30 星期三 所属： R语言代写 浏览：64

r 计量经济分析代写 This could be a group project with a maximum of three students. Write a 4-6 (typed) page (regular font size, etc!) report

This could be a group project with a maximum of three students. Do not use the same data as someone else. Write a 4-6 (typed) page (regular font size, etc!) report providing a statistical/econometric analysis and forecasts for a set of variables of your choice (your ”focus variables”).

The following conditions must be met:

1. Use at least 3 related time series variables, with at least 200 observations (per variable).
2. Carefully document the source of your data, the variable definitions, etc.
3. Your report should be clear, concise, and communicate effectively. It should not require any detective work on the part of the reader to figure out what you have done. Everything must be explained, at least briefly (but don’t write a textbook either!). Be sure to analyze your results in writing and draw conclusions.
4. Be sure to include a title, your name, the date (month, year), and page numbers.

Term Project Assignment

(a) Stationarity: Determine the order of integration of each variable.

(b) Dynamic specification: Build a dynamic regression model, choosing one of your variables as the dependent variable. Be sure to include an appropriate number of lags of all the variables, etc.

Explain how you chose your best model. Write down the equation (not code) representing your best model, as well as report the regression summary results, etc.

(c) Seasonality: Include seasonal dummy variables to check for any evidence of seasonality. If the joint F-test for the seasonal dummy variables is statistically significant (10% level at least), then keep the seasonal dummies in your model. Provide the evidence on seasonality. Try seasonal lags also and attempt to determine with dummies or lags is better, or both are needed.

(d) Serial correlation: test your best model for serial correlation in the residuals and, if detected, adjust the dynamic specification to remove any serial correlation.

(e) Granger causality: Test all lags of each variable for joint significance, i.e. test whether each variable has any effect on the other variables over time.

(f) Conclusions: Summarize your findings with respect to the relationship between the dependent and explanatory variables. Be sure to indicate whether you think that your model performs satisfactorily, etc.