Data Transformation for Trading Models

Download data from GitHub into Pandas dataframe

Student will create a Pandas dataframe that has the CSV imported from GitHub.

Check for missing data

Student will run appropriate functions on the DataFrame containing the data and report the number missing per column/attribute

Forward fill missing data

Student will use the information from the prior step (checking for missing data) to fill in any missing data in columns that require it. The should be using forward fill.

Remove special characters and convert to numeric/datetime

The student will fill in the convert_dollar_columns_to_numeric function using code that replaces dollar signs with empty strings (‘’).

Align datetime data

The student will change the datetime attribute for the inflation DataFrame so that it falls on the last day of the month.

Upsample, downsample and interpolate data

There are two separate steps here. The first creates quarterly average data and the second interpolates weekly data from monthly data.

Normalize/standardize a feature

Students will create a new attribute that holds the standardized GDP values. The solution does this manually by calculating the min and max and using it in the formula. However, it’s also acceptable if students use the StandardScaler method from sklearn or some other similar method.

Plotting a time series of adjusted open vs close price

The student will plot a line chart that displays the open and close price for Apple stock for the last three months.

Plotting a histogram of a stock's closing price in the last three months

The student will again use the last 3 months of data to plot a histogram of Apple closing price.

Calculating correlation between a stock price and a macroeconomic variable and plotting the output as a heatmap

The student will plot a heatmap of correlations between Microsoft and Apple returns and inflation. Note that this will be a 3x3 heatmap matrix as it will include the pairwise correlations of Apple, Microsoft, and inflation.

Calculating rolling volatility (standard deviation) of a stock's price for last 3 months and plotting as a line chart

The student will calculate the rolling standard deviation for Apple stock. They will then plot the rolling SD and the Apple closing price on the same chart, but using two y-axes.

Calculating rolling volatility (standard deviation) of a stock's price for last 3 months and plotting as a line chart

The student will calculate the rolling standard deviation for Apple stock. They will then plot the rolling SD and the Apple closing price on the same chart, but using two y-axes.

Export transformed data as a CSV back to local computing environment

The student will create an output CSV of the transformed data. Depending on whether they merged DataFrames in the prior steps, they may be outputting one or more CSV files in this step.