Learning a new programming language is costly. Usually it takes a considerable amount of time to get acquainted with a new language. Especially the first phase can be painful and frustrating. The good thing is that with enough time and effort most of us will learn how to master a programming language eventually. However, note that, once we are comfortable with one language, we hardly want to change again. It turns out that the cost of abandoning on programming language and switch to another are even higher than at the beginning. Knowing this, we really want to make sure not to invest in the wrong language. There might be nothing worse than after finally mastering a programming language, recognizing that there is no use for this language anymore. While in a former post I highlighted reason why to use R, I concentrate on the Pros and Cons of R in this post.
Advantages of R
Why should you use R?
There exists several reasons why one should start using R. During the last decade R has become the leading tool for statistics, data analysis, and machine learning. By now, R represents a viable alternative to traditional statistical programs such as Stata, SPSS, SAS, and Matlab. The reasons for R’s success are manifold. Continue reading
In this post I am going to explain how to enable GUI root access on Debian 9. It is very similar to enabling Gui Root Login in Debian 8. At this point I should warn you that using the root account is dangerous as you can ruin your whole system. Try to follow this guide exactly.
What is seasonal adjustment?
Seasonal adjustment refers to a statistical technique that tries to quantify and remove the influences of predictable seasonal patterns to reveal nonseasonal changes in data that would otherwise be overshadowed by the seasonal differences. Seasonal adjustments provide a Continue reading
The easiest way to compute clustered standard errors in R is the modified
summary(). I added an additional parameter, called
cluster, to the conventional
summary() function. This parameter allows to specify a variable that defines the group / cluster in your data. The summary output will return clustered standard errors. Here is the syntax:
summary(lm.object, cluster=c("variable")) Continue reading
The following R script creates an example dataset to illustrate the application of clustered standard errors. You can download the dataset here.
The script creates a dataset with a specific number of student test results. Individual students are identified via the variable
student_id . The variable
id_score comprises a student’s test score. In the test, students can score from 1 to 10 with 10 being the highest score possible. Continue reading
Clustered standard errors are a way to obtain unbiased standard errors of OLS coefficients under a specific kind of heteroscedasticity. Recall that the presence of heteroscedasticity violates the Gauss Markov assumptions that are necessary to render OLS the best linear unbiased estimator (BLUE).
The estimation of clustered standard errors is justified if there are several different covariance structures within your data sample that vary by a certain characteristic – a “cluster”. Furthermore, the covariance structures must be homoskedastic within each cluster. In this case clustered standard errors provide unbiased standard errors estimates. Continue reading