In this post, we are going to talk about cointegration. It is a concept closely related to the unit root process in a multivariate model setting. What is cointegration? Suppose that y_{1,t}, y_{2,t}, y_{3,t} · · · y_{k,t} are a set of I(1) variables. …

In this series, we have spent a lot of time talking about ARIMA models, which are essentially univariate models. In this post, we discuss the multivariate models. From Univariate to Multivariate Models Honestly, when I saw the univariate models for the first time, I had huge confusion. Why do people…

In this post, we will talk about how to identify the model and evaluate them. The full notebook can be found here. Let’s first take a look at the framework and I will explain them in detail. Check for Stationary

In this post, let’s talk about the non-stationary process. Most of the time, when we do an empirical analysis of time series, the first thing that we want to check is whether the time series data is stationary or not. Non-stationary Process If you forget what stationary means, please check…

In this post, let’s talk about the AutoRegressive Models. The derivations are from a lecture note compiled by myself, so there should not be a concern about the copyright. You can find the full note here. Motivations for AR models As we introduced in EP01, we need some structure in…

In this post, let’s talk about the Moving Average Model. The derivations are from a lecture note compiled by myself, so there should not be a concern about the copyright. You can find the full note here. Motivations for MA models As we introduced in EP01, we need some structure…

In this post, I will discuss several essential concepts for time series forecasting. This post is designed to be a reference for later posts, so the sections are not connected closely. You can use “ctrl+F” for efficient search. Also, though this post is intended to explain some technical details, it…

I developed a new series of posts to introduce the concepts of forecasting in Economics. These posts are not as rigorous as a lecture that explains every technical detail. The main purpose is to help readers to build connections between scattered key concepts. In this post, I will briefly introduce…

In this post, let’s talk about power analysis in a bit more detail. 1. What does Statistical Power mean? Power is the probability of detecting an effect, given that the effect is really there. In other words, it is the probability of rejecting the null hypothesis when it is in…