Compute portfolio average long-term rating¶

The following case-study demonstrates how to compute the average long-term rating of your portfolio.

Preliminary tasks¶

As a first step, we are going to import a portfolio into a pd.DataFrame. We'll call it port_df. This dataframe comprises a number of securities with respective weights and ratings.

In [1]:

import pandas as pd

import pyratings as rtg

port_df = pd.read_excel("portfolio.xlsx", sheet_name="long_term_ratings_worst")
port_df.head()

import pandas as pd import pyratings as rtg port_df = pd.read_excel("portfolio.xlsx", sheet_name="long_term_ratings_worst") port_df.head()

Out[1]:

	ISIN	weight	worst_rtg
0	ISIN00000001	0.518515	AAA
1	ISIN00000002	0.950810	AA+
2	ISIN00000003	0.497176	AA
3	ISIN00000004	0.648453	AA-
4	ISIN00000005	0.674328	AA-

Translating ratings into numerical scores¶

The column worst_rtg will be used in order to translate the ratings into scores according to the following table.

We will use get_scores_from_ratings to translate the human-readable ratings into numerical scores. The function needs a rating provider (here: "S&P") in order to select an appropriate translation dictionary.

In [2]:

port_scores_df = pd.concat(
    [
        port_df,
        rtg.get_scores_from_ratings(
            ratings=port_df["worst_rtg"],
            rating_provider="S&P")
    ],
    axis=1
)

port_scores_df

port_scores_df = pd.concat( [ port_df, rtg.get_scores_from_ratings( ratings=port_df["worst_rtg"], rating_provider="S&P") ], axis=1 ) port_scores_df

Out[2]:

	ISIN	weight	worst_rtg	rtg_score_worst_rtg
0	ISIN00000001	0.518515	AAA	1.0
1	ISIN00000002	0.950810	AA+	2.0
2	ISIN00000003	0.497176	AA	3.0
3	ISIN00000004	0.648453	AA-	4.0
4	ISIN00000005	0.674328	AA-	4.0
...	...	...	...	...
82	ISIN00000083	2.321185	BBB+	8.0
83	ISIN00000084	1.389043	BBB+	8.0
84	ISIN00000085	2.296711	BBB+	8.0
85	ISIN00000086	1.015105	AAA	1.0
86	ISIN00000087	0.964399	AA	3.0

87 rows × 4 columns

Here, the get_scores_from_ratings function has been used to translate the worst_rtg column into numerical scores. That is, we used a column that has already been consolidated.

pyratings offers some convenience functions that consolidate multiple ratings and directly compute the respective numerical rating scores. These functions are

• get_best_scores
• get_second_best_scores
• get_worst_scores

Finally, we need to compute a weighted average rating score, which we subsequently convert back into a human-readable rating. For the former, we use the get_weighted_average function and for the latter the get_ratings_from_scores function.

In [3]:

avg_rtg_score = rtg.get_weighted_average(
    data=port_scores_df["rtg_score_worst_rtg"],
    weights=port_scores_df["weight"] / 100,
)

print(f"Average rating score: {avg_rtg_score:.2f}")

avg_rtg_score = rtg.get_weighted_average( data=port_scores_df["rtg_score_worst_rtg"], weights=port_scores_df["weight"] / 100, ) print(f"Average rating score: {avg_rtg_score:.2f}")

Average rating score: 4.39

In [4]:

avg_rtg = rtg.get_ratings_from_scores(
    rating_scores=avg_rtg_score,
    rating_provider="S&P"
)

print(f"Average portfolio rating: {avg_rtg}")

avg_rtg = rtg.get_ratings_from_scores( rating_scores=avg_rtg_score, rating_provider="S&P" ) print(f"Average portfolio rating: {avg_rtg}")

Average portfolio rating: AA-

There you go. The average rating of the portfolio is AA-.

Do you prefer to show the rating using Moody's rating scale? — It's as simple as changing the rating_provider argument to "Moody".

In [5]:

avg_rtg = rtg.get_ratings_from_scores(
    rating_scores=avg_rtg_score,
    rating_provider="Moody"
)

print(f"Average portfolio rating: {avg_rtg}")

avg_rtg = rtg.get_ratings_from_scores( rating_scores=avg_rtg_score, rating_provider="Moody" ) print(f"Average portfolio rating: {avg_rtg}")

Average portfolio rating: Aa3