Research Case Study Template

import talib as ta

#Get Data
df = get_pricing('spy' , start_date='2003-01-01' , end_date='2019-09-16' , frequency='daily')

#Get Trailing 1,3,5 Day Percent Changes, drop NAs and observe the baseline results
df['percent_ch_1d'] = df['close_price'].pct_change(1)
df['percent_ch_3d'] = df['close_price'].pct_change(3)
df['percent_ch_5d'] = df['close_price'].pct_change(5)
df.dropna(inplace=True)
df[['percent_ch_1d','percent_ch_3d','percent_ch_5d']].describe()

#Calc RSI
df['rsi_4'] = ta.RSI(df['close_price'] , 4)

#Shift closes back in time
df['FUT_1d_c'] = df['close_price'].shift(-1)
df['FUT_3d_c'] = df['close_price'].shift(-3)
df['FUT_5d_c'] = df['close_price'].shift(-5)

#Calculate future percent returns and drop NAs
df['FUT_1d_perc'] = df['FUT_1d_c'].pct_change(1)
df['FUT_3d_perc'] = df['FUT_3d_c'].pct_change(3)
df['FUT_5d_perc'] = df['FUT_5d_c'].pct_change(5)
df.dropna(inplace=True)

#Make a Condition
condition_1 = df['rsi_4'] < 30
filtered_df = df[condition_1]

#See future results
filtered_df[['FUT_1d_perc','FUT_3d_perc','FUT_5d_perc']].describe()