Trading Backtesting function
Backtesting is the process of applying a trading strategy or analytical method to historical data to see how accurately the strategy or method would have predicted actual results.
Included now into the T library: https://github.com/al3xandr3/T
Basic Usage
Lets use the S&P500 index as the stock to invest in
>>> sp500 = t.get_quotes_close(['^GSPC'], date_from = '2018-06-14')
>>> sp500 = sp500.reset_index()
Lets say we have 1000 to invest
>>> capital = 1000
What are the gain/losses if we had invested 1000 in S&P in 2018-06-14 ?
>>> bulk1 = pd.DataFrame()
>>> bulk1["date"] = sp500["date"]
>>> bulk1["order_size"] = 0
>>> bulk1.loc[t.where(bulk1, "date", datetime.strptime('2018-06-14', '%Y-%m-%d'), op.eq).index[0], "order_size"] = 1000
>>> out1 = t.backtest_strategy(sp500, bulk1, capital)
>>> out1["transactions"]
| date | ^GSPC | order_size | pct_change | invested_start_day | invested_end_day | account_cash_start_day | account_cash_end_day | net_worth | nb | |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 2018-06-14 | 2782.489990 | 1000 | NaN | 0.000000 | 1000.000000 | 1000 | 0 | 1000.000000 | |
| 1 | 2018-06-15 | 2779.659912 | 0 | -0.001017 | 998.982897 | 998.982897 | 0 | 0 | 998.982897 | |
| 2 | 2018-06-18 | 2773.750000 | 0 | -0.002126 | 996.858932 | 996.858932 | 0 | 0 | 996.858932 | |
| 3 | 2018-06-19 | 2762.590088 | 0 | -0.004023 | 992.848167 | 992.848167 | 0 | 0 | 992.848167 | |
| 4 | 2018-06-20 | 2767.320068 | 0 | 0.001712 | 994.548077 | 994.548077 | 0 | 0 | 994.548077 | |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 495 | 2020-06-03 | 3122.870117 | 0 | 0.013649 | 1122.329327 | 1122.329327 | 0 | 0 | 1122.329327 | |
| 496 | 2020-06-04 | 3112.350098 | 0 | -0.003369 | 1118.548533 | 1118.548533 | 0 | 0 | 1118.548533 | |
| 497 | 2020-06-05 | 3193.929932 | 0 | 0.026212 | 1147.867537 | 1147.867537 | 0 | 0 | 1147.867537 | |
| 498 | 2020-06-08 | 3232.389893 | 0 | 0.012042 | 1161.689675 | 1161.689675 | 0 | 0 | 1161.689675 | |
| 499 | 2020-06-09 | 3207.179932 | 0 | -0.007799 | 1152.629459 | 1152.629459 | 0 | 0 | 1152.629459 |
And the overall percent gain/loss
>>> out1["lift"]
0.153
about 15% gain, i.e. +152 gain from the original 1000 investment.