vindex - 7 months ago 37
Python Question

# Python: for loop to find how many stocks hits 52 weeks high and low

I can calculate how many stocks that is in the 52 weeks new high or new low for the last trading day. But I need to calculate from the first day in the csv file till the last day in the csv.

Example:

02-01-2014 , 10 stocks 52 weeks high and 45 is 52 weeks low

03-01-2014, 23 stocks 52 weeks high and 56 stocks 52 weeks low

04-01-2014, 34 stocks 52 weeks high and 34 stocks 52 weeks low.

``````import pandas as pd
import numpy as np
import csv
import datetime
import matplotlib.pyplot as plt
import talib as ta
import stocklist

now = datetime.datetime.now()

STOCKS = ['Abc','cdf','gg','D','AN','OX']
Stockslen = len(STOCKS)

h_cnt=0
l_cnt=0

#Creating 5 df for data analysis

df_today52w_High = pd.DataFrame(columns=['Stock','Today 52w_High'])
df_today52w_Low = pd.DataFrame(columns=['Stock','Today 52w_Low'])

for x in range (len(STOCKS)):
print "###############  "
print STOCKS [x]
print "###############"
q_data = pd.read_csv(STOCKS [x]+".csv", index_col='Stock', usecols =[0,1,3,4,5,6,7])

high = q_data.High
h=np.array(high)

date_ = q_data.Date
dt = np.array(date_)

open_ = q_data.Open
o = np.array(open_)

low = q_data.Low
l = np.array(low)

close = q_data.Close
c = np.array(close)

if h[-1] == ta.MAX(h,252)[-1]:
df_today52w_High.loc[len(df_today52w_High)] = [STOCKS[x],1]
h_cnt += 1
print h_cnt
else:
df_today52w_High.loc[len(df_today52w_High)] = [STOCKS[x],0]

if l[-1] == ta.MIN(l,252)[-1]:
df_today52w_Low.loc[len(df_today52w_Low)] = [STOCKS[x],1]
l_cnt += 1
print l_cnt
else:
df_today52w_Low.loc[len(df_today52w_Low)] = [STOCKS[x],0]

df_new = pd.merge(df_today52w_High,df_today52w_Low,how='outer',on='Stock')

df_new['52w high']= h_cnt
df_new['52w low']= l_cnt
``````

The csv in the STOCKS has format as below. I have 300 stocks in the STOCKS list. I just shows a few here.

``````Stock,Date,Time,Open,High,Low,Close,Volume
AAX,2014-01-02,00:00:00,1.0,1.02,1.0,1.01,3251900
AAX,2014-01-03,00:00:00,1.01,1.05,1.01,1.03,8416100
AAX,2014-01-06,00:00:00,1.04,1.05,1.02,1.03,2625200
AAX,2014-01-07,00:00:00,1.03,1.03,1.01,1.01,2539700
AAX,2014-01-08,00:00:00,1.02,1.02,1.0,1.02,2072700
AAX,2014-01-09,00:00:00,1.02,1.02,1.0,1.01,2589600
AAX,2014-01-10,00:00:00,1.01,1.01,1.0,1.01,2057200
AAX,2014-01-13,00:00:00,1.01,1.01,1.0,1.0,1284000
AAX,2014-01-15,00:00:00,1.0,1.01,1.0,1.0,1938100
.
.
AAX,2016-02-29,00:00:00,0.25,0.26,0.24,0.25,63660600
AAX,2016-03-01,00:00:00,0.25,0.26,0.25,0.26,100823200
AAX,2016-03-02,00:00:00,0.27,0.28,0.26,0.28,57543300
AAX,2016-03-03,00:00:00,0.28,0.29,0.27,0.28,113837600
AAX,2016-03-04,00:00:00,0.29,0.3,0.28,0.3,138182600
``````

Instead of df using `writerow`

``````if h[y]== ta.MAX(h,20)[y]:
csvout = open('52w_h.csv', 'a')
csvwrite = csv.writer(csvout)
csvwrite.writerow([STOCKS [x][0]]+[dt[y]]+["1"])
csvout.close()
else:
csvout = open('52w_h.csv', 'a')
csvwrite = csv.writer(csvout)
csvwrite.writerow([STOCKS [x][0]]+[dt[y]]+["0"])
csvout.close()
``````

Then u can group the date using `groupby`

``````a = pd.read_csv("52w_h.csv")
b = a.groupby('Date')
df_h= b['52wh'].sum()
``````

Output:

``````2016-04-06 160
2016-04-07 170
2016-04-08 142
``````