The Stock Market Works by Day, but It Loves the Night
The daytime is for losers. Overnight is when the big money is made in the stock market — not by trading but by getting a good night’s sleep.
def point_plot(df, ax):
x = df.iloc[-1:].reset_index()
x.plot.scatter('Date', 'intraday_cum_returns', ax = ax, s = 150, color = 'red', zorder = 10)
x.plot.scatter('Date', 'overnight_cum_returns', ax = ax, s = 150, color = '#444444', zorder= 10 )
ax.text(x['Date'] + td(days = 180) , x['intraday_cum_returns']-0.3, "Regular\nHours", fontsize = 14)
ax.text(x['Date'] + td(days = 180) , x['overnight_cum_returns']-0.3, "Outside Regular Hours", fontsize = 14)
return df
fig, ax = plt.subplots(figsize=(12,9))
.5
(yf
.Ticker("SPY")
.history(period = 'max', back_adjust=True)
.assign(intraday_pct_change = lambda x: (x['Close'] - x['Open']) / x['Open'],
intraday_cum_returns = lambda x: (1 + x['intraday_pct_change']).cumprod() - 1,
overnight_pct_change = lambda x: (x['Open'] - x['Close'].shift(1)) / x['Close'].shift(1),
overnight_cum_returns= lambda x: (1 + x['overnight_pct_change']).cumprod() - 1
)
[['intraday_cum_returns',
'overnight_cum_returns']]
.pipe(point_plot, ax)
.plot(logy=False, ax=ax, color = ['red','#444444'], legend = False)
)
plt.grid(axis= 'y')
[ax.spines[i].set_visible(False) for i in ['top','right','left','bottom']]
plt.ylabel("")
plt.yticks(range(0,18,2))
ax.set_yticklabels(['{:.0%}'.format(x) for x in ax.get_yticks()])
plt.text(datetime(1989,1,1), 20, "Profiting After the Market Closes", font = "Georgia" ,size = 25, fontweight='bold')
plt.text(datetime(1989,1,1), 18, "Since 1993, all of the S.& P. 500's gains have occurred outside regular trading hours, which run\nfrom 9:30 a.m. to 4 p.m. Eastern time.", font = "Serif" ,size = 15, fontweight='normal', color = '#444444')
plt.text(datetime(1989,1,1), 17, "@MichaelHotaling")
def point_plot(df, ax):
x = df.iloc[-1:].reset_index()
x.plot.scatter('Date', 'intraday_cum_returns', ax = ax, s = 150, color = 'red', zorder = 10)
x.plot.scatter('Date', 'overnight_cum_returns', ax = ax, s = 150, color = '#444444', zorder= 10)
ax.text(x['Date'] + td(days = 180) , x['intraday_cum_returns']-0.3, "Regular\nHours", fontsize = 14)
ax.text(x['Date'] + td(days = 180) , x['overnight_cum_returns']-0.3, "Outside Regular Hours", fontsize = 14)
return df
fig, ax = plt.subplots(figsize=(12,9))
.5
(yf
.Ticker("SPY")
.history(period = 'max', back_adjust=True)
.assign(intraday_pct_change = lambda x: (x['Close'] - x['Open']) / x['Open'],
intraday_cum_returns = lambda x: (1 + x['intraday_pct_change']).cumprod(),
overnight_pct_change = lambda x: (x['Open'] - x['Close'].shift(1)) / x['Close'].shift(1),
overnight_cum_returns= lambda x: (1 + x['overnight_pct_change']).cumprod()
)
[['intraday_cum_returns',
'overnight_cum_returns']]
.pipe(point_plot, ax)
.plot(logy=True, ax=ax, color = ['red','#444444'], legend = False)
)
ax.grid(axis="y", )
[ax.spines[i].set_visible(False) for i in ['top','right','left','bottom']]
plt.ylabel("")
#plt.yticks(range(0,18,2))
ax.set_yticks([2**x for x in range(-1,5)])
ax.set_yticklabels(['{:.0%}'.format(x-1) for x in [2**x for x in range(-1,5)]])
plt.text(datetime(1989,1,1), 37, "Profiting After the Market Closes", font = "Georgia" ,size = 25, fontweight='bold')
plt.text(datetime(1989,1,1), 25, "Since 1993, all of the S.& P. 500's gains have occurred outside regular trading hours, which run\nfrom 9:30 a.m. to 4 p.m. Eastern time.", font = "Serif" ,size = 15, fontweight='normal', color = '#444444')
plt.text(datetime(1989,1,1), 19, "@MichaelHotaling")
