Session 4: Performance measurement

Maximum drawdown

import os 
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt 
import warnings

warnings.simplefilter('ignore')

#os.chdir(r'C:\Users\cdes1\OneDrive - ICHEC\Documents\Cours\PortfolioManagement')
os.chdir(r'/Users/christophe/OneDrive - ICHEC/Documents/Cours/PortfolioManagement')

df = pd.read_csv('data/SP500.csv', index_col=0, parse_dates=True)

rets = df.pct_change().dropna()
dates = df.index[1:]
# convert to numpy  
rets = np.array(rets[rets.columns[0]])
drawdown = np.zeros_like(rets)
max_DD = np.zeros_like(rets)
to_through = np.zeros_like(rets)
recovery = np.zeros_like(rets) 

runs = []

# initialisation at first date 
drawdown[0] = rets[0]
max_DD[0] = rets[0]
to_through[0] = 1 

length = 1
to_through = 1
start = dates[0]
start_t = 0 
bottom = dates[0]
bottom_t = 0 

for t in range(1, len(rets)):
    cum_r = (1+drawdown[t-1])*(1+rets[t])-1
    drawdown[t] = np.minimum(0, cum_r)
    # on continue la serie 
    if (drawdown[t] < 0) and (t < len(rets)-1):   
        length = length + 1
        if max_DD[t-1] < drawdown[t]:
            max_DD[t] = max_DD[t-1]
        else:
            max_DD[t] = drawdown[t] 
            bottom = dates[t]
            bottom_t = t
        
    else: # on recommence une serie
        # on stocke toutes les infos
        to_through = bottom_t - start_t
        recovery = t - bottom_t 
        end = dates[t]
        if t == len(rets)-1:
            end = np.NAN # we have not recover yet 
        infos = {'start': start,
                 'bottom': bottom,
                 'end': end,
                 'to_through': to_through,
                 'recovery': recovery,
                 'nbDays': length,
                 'maxDD': max_DD[t-1]}
        runs.append(infos)
        # puis on reinitialise les variables  
        length = 1
        start = dates[t]
        start_t = t
        bottom_t = t
        
df_dd = pd.DataFrame(data=drawdown, index=dates, columns=['Drawdown'])
fig, ax = plt.subplots(figsize=(10, 6))
ax.plot(df_dd['Drawdown'])
vals = ax.get_yticks()
ax.set_yticklabels(['{:,.2%}'.format(x) for x in vals]);

df_runs = pd.DataFrame(runs)
df_runs.sort_values(by='maxDD', ascending=True, inplace=True)
df_runs.head(n=10)

	start	bottom	end	to_through	recovery	nbDays	maxDD
531	2007-10-09	2009-03-09	2013-03-28	355	1021	1376	-0.567754
522	2000-03-24	2002-10-09	2007-05-30	637	1166	1803	-0.491469
786	2020-02-19	2020-03-23	2020-08-18	23	103	126	-0.339250
195	1987-08-25	1987-12-04	1989-07-26	71	414	485	-0.335095
42	1980-11-28	1982-08-12	1982-11-03	430	58	488	-0.271136
877	2022-01-03	2022-10-12	NaT	195	92	287	-0.254251
214	1990-07-16	1990-10-11	1991-02-13	62	86	148	-0.199187
738	2018-09-20	2018-12-24	2019-04-23	65	81	146	-0.197782
477	1998-07-17	1998-08-31	1998-11-23	31	59	90	-0.193360
17	1980-02-13	1980-03-27	1980-07-14	30	74	104	-0.170719