Solution

df["Group"] = model.labels_
print(df)

We are going to want to look at different groups and see what they look like. First, I want to show you how to get a truth series that represents all the stocks in a group.

print(df["Group"]==1)

You index by it the same way as before.

print(df[df["Group"]==1])

Let’s loop through each group and see what stocks are in it.

for x in range(8):
    print(df[df["Group"]==x])

Our next step is going to be bringing in the full table we downloaded from wikipedia and using it so that we can get information about each group’s industries.

SP500 = pd.DataFrame.from_csv("SP500.csv", encoding="UTF-8")
tickers = []
for x in SP500["Ticker"].values:
    if "." in x:
        x = x.replace(".","")
    tickers.append(x)
SP500["Ticker"]=tickers

The extra code just ensures we have the correct version of each ticker so the two datasets will match.