`G0121: Matrix not positive definite` and `G0048: Matrix singular` are common errors encountered during estimation. Today we will learn how to diagnose these errors using GAUSS code to compute ordinary least squares estimates, using real data from some golf shots hit by this author and recorded by a launch monitor.
Last week we learned how to use the `date` keyword to load dates into GAUSS. Today, we extend our analysis of time series data to plot high-frequency Forex data.
Time series data with inconsistently formatted dates and times can make your work frustrating. Dates and times are often stored as strings or text data and converting to a consistent, numeric format might seem like a daunting task. Fortunately, GAUSS includes an easy tool for loading and converting dates and times – the `date` keyword.
If you have run much publicly available GAUSS code, you have probably come across the `#include` command. In this blog, we answer some important questions about #include:
What does `#include` do?
What is the most common error when using `#include`?
In this blog, we explore data path best practices for making GAUSS code more portable and replicable. Using variables and predefined GAUSS path definitions, we show how to simplify code and easily customize data loading.
Autocomplete is becoming a common feature in the tools we use in all aspects of our lives, because of it’s ability to help us to type more accurately and quickly. When programming in GAUSS, the autocomplete can also show you new functions you were not aware of. Today we will discuss how to use and control autocomplete features of the GAUSS editor and command window.
The key to getting the most performance from a matrix language is to vectorize your code as much as possible. Vectorized code performs operations on large sections of matrices and vectors in a single operation, rather than looping over the elements one-by-one. In this blog, we learn how to use the GAUSS recserar function to vectorize code and simulate a time series AR(1) model.
Starting in GAUSS version 12, a new suite of high quality and high-performance random number generators was introduced. While new projects should always use one of the modern RNG’s, it is sometimes necessary to exactly reproduce some work from the past. GAUSS has retained a set of older LCG’s, which will allow you to reproduce the random numbers from older GAUSS versions for many distributions.
Many estimations and forecasting methods are not valid if the mean and variance are not constant across time. Today we examine how to test for both using GLS-unit root tests with multiple structural breaks.
While structural breaks are a widely examined topic in pure time series, their impacts on panel data models have garnished less attention.
However, in their forthcoming paper Chowdhury and Russell (2018)] demonstrate that structural breaks can cause bias in the instrumental variable panel estimation framework.
This work highlights that structural breaks shouldn’t be limited to pure time series models and warrant equal attention in panel data models.
