Dealing with Seasonality in R Part 1 - Exploratory Data Analysis and Data Cleaning

This is the first video of a series on dealing with seasonality in R. This is a complete walkthrough and will show you how to identify and account for seasonality, trending and more. The dataset is kratom sales from a local head shop.

In this video we primarily deal with exploratory data analysis (EPA) and data cleaning in r. We load the libraries and then take the data and place it in a data frame with the data.frame r function. We do this because we need the added functionality that we will not get in a tibble.

Get a free stock like Apple, Ford, or Facebook worth up to $500 just by signing up and joining Robinhood using my link: https://join.robinhood.com/davidm16707

Next we concatenate two fields into one for ordering simplicity. I then show you an order function with indices that makes the ordering process simple as running a line of code. Then we turn the data into a time series object and cleanse it.

We then get the weekly and monthly moving averages and then check with the summary function for NA's or missing values. We then use na.aggregate to replace the missing values with the column mean and graph (with the plot function and ggplot library) both the pre and post data to show the difference.

This video mostly deals with exploratory data analysis. In the next video we will deal with decomposing the data and identifying and graphing the seasonality, trends, etc... Again, this is part 1 of this seasonality analysis. We are building a reusable predictive arima r forecasting model that will yield properly deseasonalized data and predictive models with very high accuracy. We will also test and graph each part numerous times so we can rely strongly on our results.

I hope you found this interesting and helpful.

Please take a moment to like and subscribe and be sure to share!

Thanks again and God Bless!