Python & R: Merging Time Series Data

Rolling Forward Fill Merges in R and Python, and why you should do it too. Can also just use the function we build in there.

Required Readings:

Time Series analysis & Lookahead Bias

R & Python Programming (libraries)

Data Wrangling

Table of Contents:

1. Introduction

2. Rolling Forward Data (Re-sampling)

3. Merging Data

4. Rolling Merges

1 - Introduction

You can download the GAP data here:
You can download the Unemployment data here:
You can download the CPI data here:

Sometimes, we will have to pull chunks of data from multiple locations, and then merge them together to make a new dataset. There are a few different ways to do it, this post will focus on: merging data, doing forward fills, and doing them together.

Hence the emphasis on time series merging.

2 - Rolling Forward Data (Re-sampling)

Notice the FRED economic data is in a monthly format, but the stock data is in a daily format. Now, since I just grabbed the data directly from FRED instead of using the app I developed, to make it fair, I need to roll the FRED data forward about 1 month (at least). In other words the Date Jan 1, 2021 becomes Jan 31, 2021.

This is also known as re-sampling the data.

The stock data on the other hand is available on a daily basis, so we won’t need to roll that one forward, and can just keep it as is.

Python

To handle this one easily, we will create a new column called Ceiling_Date. To