Business Context
A financial investment firm needs to assess and compare the risk profile of different market sectors. The goal is to analyze the daily return volatility of representative stocks within each sector to inform asset allocation strategies. A wider distribution and more outliers suggest higher risk.
About the Set : percentile
Precise calculation of percentiles and quantiles.
Discover all actions of percentile Data Preparation
Creation of a simulated dataset representing daily stock returns for various stocks across three market sectors: Technology, Healthcare, and Financials.
Copied!
| 1 | DATA casuser.stock_returns; |
| 2 | call streaminit(123); |
| 3 | DO sector_id = 1 to 3; |
| 4 | IF sector_id = 1 THEN sector = 'Technology'; |
| 5 | ELSE IF sector_id = 2 THEN sector = 'Healthcare'; |
| 6 | ELSE sector = 'Financials'; |
| 7 | DO stock_id = 1 to 50; |
| 8 | DO day = 1 to 252; /* Trading days in a year */ |
| 9 | IF sector = 'Technology' THEN daily_return = (rand('NORMAL', 0.05, 0.8)); |
| 10 | ELSE IF sector = 'Healthcare' THEN daily_return = (rand('NORMAL', 0.02, 0.4)); |
| 11 | ELSE daily_return = (rand('NORMAL', 0.03, 0.6)); |
| 12 | OUTPUT; |
| 13 | END; |
| 14 | END; |
| 15 | END; |
| 16 | keep sector stock_id day daily_return; |
| 17 | RUN; |
Étapes de réalisation
1
Load the simulated stock returns data into an in-memory CAS table.
Copied!
| 1 | |
| 2 | PROC CASUTIL; |
| 3 | load |
| 4 | DATA=casuser.stock_returns outcaslib='casuser' casout='stock_returns' replace; |
| 5 | QUIT; |
| 6 |
2
Execute the boxPlot action, grouping by the 'sector' variable to calculate distribution statistics for the 'daily_return' of each sector.
Copied!
| 1 | PROC CAS; |
| 2 | percentile.boxPlot / |
| 3 | TABLE={name='stock_returns', groupBy={'sector'}}, |
| 4 | inputs={{name='daily_return'}}; |
| 5 | RUN; |
Expected Result
The action returns a result table containing distinct box plot statistics (Q1, median, Q3, min, max) for the 'daily_return' variable for each market sector. The 'Technology' sector is expected to show a wider interquartile range and higher whisker values, confirming higher volatility compared to 'Healthcare'.