Economics - R Studio - Integrated Development Environment - UK M3 - Report Writing Assessment Answer

Internal Code: 1AFHAC

R Studio Integrated Development Environment Assessment Answer

Assignment Task:  R Studio Integrated Development Environment - This course project is to build the ARMA model for the time series of UK M3 (one measure of money supply) and evaluate the forecasting performance. The econometric analysis should be implemented by the statistical software R and its integrated development environment (IDE), R studio. The project report should be written in the style of academic article, including abstract, introduction, econometric analysis (the econometrics tasks), conclusion, and references (in Harvard style). For the tasks, you need to not only report the econometric results, but also provide the technical description on the methods and the interpretation/discussion on the econometric results. For example, you should include the following content for the task of ADF test:

• technical description of ADF test;
• formulas of ADF test;
• R code to conduct the analysis;
• R output (preferred to be nicely formatted tables, rather than the raw screenshot from R studio);
• how to interpret the results;
• discuss the results.

DATA: The file “UK M3.xls” contains the seasonally adjusted quarterly UK M3 data from 1987 Q1 to 2018Q4, in total 128 observations 1 . The data were obtained from the Federal Reserve Bank of St. Louis (https://fred.stlouisfed.org/series/MABMM301GBQ189S). The variables that we are interested in are log(M3t) and the growth rate, i.e. ? log(M3t) = log(M3t) ? log(M3t?1). ECONOMETRIC TASKS:

• [Pre-analysis] Provide the data visualizations (time series plot, box plot, histogram, and ACF plot) and the descriptive statistics of log(M3t) and ? log(M3t). What can you spot from various plots?

What is the meaning of descriptive statistics?

• [Unit root/stationarity tests] Perform ADF test, PP test, and KPSS test on log(M3t) and ? log(M3t). You can use AIC to select the best lag length for the ADF test and use “long” lag length for the PP and KPSS test. Typically, we only include a constant/drift in the three tests.
• [Model selection] Select the best ARMA(p,q) model for ? log(M3t) by AIC. You can set the possible values of p and q between 0 and 4, inclusive.

• [Model diagnostics] Perform various model diagnostics for the residuals from the best-selected model. Check the stationary condition and invertible condition of the estimated model.
• [Forecasting] Split the dataset into the in-sample period and out-of-sample period. The in-sample period is from 1987 Q2 to 2009 Q3 (90 in-sample observations). The out-of-sample period is from 2009 Q4 to 2018 Q4 (37 out-of-sample observations). Choose the best model only based on the data in the in-sample period 2. Plot the predicted values from the best model versus the true values.

Calculate the forecasting performance (MSE, MAE, MAPE, and %correct sign) of the best model.

• [Recursive Forecasting] The idea of recursive forecasting is to update the best model when there is a new observation coming in and use the updated model to make a one-step-ahead prediction. We still split the dataset in the same way as Task v. For each time-point t in the out-of-sample period, use all the data up to time-point t ? 1 and make prediction for the value at time-point t .

To be very clear, you need to:

• use data from 1987 Q2 to 2009 Q3 to select the best model, estimated the model, and make the prediction for 2009 Q4 only.

• use data from 1987 Q2 to 2009 Q4 to select the best model, estimated the model, and make the prediction for 2010 Q1 only.

• use data from 1987 Q2 to 2018 Q3 to select the best model, estimated the model, and make the prediction for 2018 Q4 only.

Collect the recursively predicted values. Plot the recursively predicted values versus the true values. Calculate the forecasting performance (MSE, MAE, MAPE, and %correct sign) of the recursive forecasting method. Compare the forecasting performance of recursive forecasting method in Task vi. and the forecasting method in Task v. [Hint: 1. write a function to produce one-step-ahead prediction based on a given dataset; 2. call that function in a for loop; 3. store the  predicted values in a container.] Tips: you need to provide the technical description of the econometric tools (include relevant formulas), R code, econometric results, and interpretation/discussion for each task above. Only providing the result is not adequate.

This Economics Assessment has been solved by our Economics experts at My Uni Paper. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+ Students in Australia, UK & US by helping them to score HD in their academics. Our Experts are well trained to follow all marking rubrics & referencing style.