Dear members of the Stata forum,
I hope this message finds you well. I am currently working on a research project focused on determining the factors influencing the creation of sovereign wealth funds (SWFs). In this regard, I am seeking your guidance on two specific aspects related to the panel logit model and dataset manipulation in Stata.
(i) Dataset Manipulation:
To address potential bias arising from one-year spikes in the data, I would like to calculate five-year moving averages of the independent variables, namely GDP, natural resource rents, foreign reserves, corruption, and the interaction of corruption and rents. The approach I aim to adopt is similar to that outlined in a referenced paper, which states:
"To avoid biased results arising from one-year spikes in the data, five-year moving averages of the variables were calculated. Averages were calculated only if at least three of the previous five years had data available. For example, the data points corresponding to Norway creating its SWF in 1990 were the averages of variable data from 1985 to 1989. As the Norwegian government made the decision to establish the SWF in 1990, including the data from 1990 would not accurately capture the lead-up period that influenced this decision. Once a country created an SWF, it was dropped from the sample. In the case of Norway, no data from 1990 onwards was included in the analyses."
Could you please guide me on how to implement this approach in Stata? Specifically, how can I create a dataset that incorporates the five-year moving averages for the independent variables while adhering to the conditions outlined in the quoted passage?
(ii) Panel Logistic Regression:
I intend to perform a panel logistic regression to examine the effect of the last five-year moving average of the independent variables (GDP, reserves, resource rents, and corruption) on the creation of SWFs, which is represented as a dummy variable. Could you kindly provide guidance on how to conduct the panel logistic regression in Stata, considering the aforementioned independent variables and their last five-year moving averages?
Your expertise and insights in addressing these matters would be immensely valuable to me. Thank you in advance for your time and assistance. I look forward to your response and the opportunity to learn from your expertise.
Best regards
I hope this message finds you well. I am currently working on a research project focused on determining the factors influencing the creation of sovereign wealth funds (SWFs). In this regard, I am seeking your guidance on two specific aspects related to the panel logit model and dataset manipulation in Stata.
(i) Dataset Manipulation:
To address potential bias arising from one-year spikes in the data, I would like to calculate five-year moving averages of the independent variables, namely GDP, natural resource rents, foreign reserves, corruption, and the interaction of corruption and rents. The approach I aim to adopt is similar to that outlined in a referenced paper, which states:
"To avoid biased results arising from one-year spikes in the data, five-year moving averages of the variables were calculated. Averages were calculated only if at least three of the previous five years had data available. For example, the data points corresponding to Norway creating its SWF in 1990 were the averages of variable data from 1985 to 1989. As the Norwegian government made the decision to establish the SWF in 1990, including the data from 1990 would not accurately capture the lead-up period that influenced this decision. Once a country created an SWF, it was dropped from the sample. In the case of Norway, no data from 1990 onwards was included in the analyses."
Could you please guide me on how to implement this approach in Stata? Specifically, how can I create a dataset that incorporates the five-year moving averages for the independent variables while adhering to the conditions outlined in the quoted passage?
(ii) Panel Logistic Regression:
I intend to perform a panel logistic regression to examine the effect of the last five-year moving average of the independent variables (GDP, reserves, resource rents, and corruption) on the creation of SWFs, which is represented as a dummy variable. Could you kindly provide guidance on how to conduct the panel logistic regression in Stata, considering the aforementioned independent variables and their last five-year moving averages?
Your expertise and insights in addressing these matters would be immensely valuable to me. Thank you in advance for your time and assistance. I look forward to your response and the opportunity to learn from your expertise.
Best regards
