Source: stocks from google finance : https://www.google.com/finance

Data: Historical prices from NASDAQ and NYSE for last 10 years. Which includes variables : Open,High,Low,Close,Volume.
10 stock sources from NASDAQ - AAPL(Apple), AMD, AMZN(Amazon),ATVI,FB(Facebook),GOOG,GOOGL(Google),INTC(Intel),
MSFT(Microsoft),TSLA(Tesla).
10 stock sources from NYSE - ABB, AFG, ALK, BAC, C, GCH, GE, IBM, JPM(JP Morgan), LVS, T.

Methods :Preprocessing the variables of the datasets. The datasets are stock sources which are loaded into the R enviroment from csv format.

• Date: Originally date format in the dataset is 16-Mar-17,and its data type is factor, which is converted to date type using as.date() to 2017-03-16. We write a function to transform date column for all the sources.

• Columns: Here we are working with 10 stock sources from NASDAQ, each stock source dataset consists of 5 columns Date,Open,High,Low,Close,Volume. Below is a descriptive view of the dataframe of one of the sources.

Listing the stocks and their individual length.

tickers= temp
j=0
for(t in tickers)
{
  j=j+1
  a =get(noquote(t))[,1]
  print(c(t,length(a)))

}  

[1] "aapl.csv" "4000"    
[1] "amd.csv" "4000"   
[1] "amzn.csv" "4000"    
[1] "atvi.csv" "4000"    
[1] "fb.csv" "1215"  
[1] "goog.csv" "3165"    
[1] "googl.csv" "3165"     
[1] "intc.csv" "4000"    
[1] "msft.csv" "4000"    
[1] "tsla.csv" "1691"    

Considering sources which has longest historical price range, based on their number of rows we select the sources having 4000 rows. We create a datasets Open,High,Low,Close . For instance Open dataset consists of open prices for all the stock sources having 4000 rows, so on for Close, High and Low.

Open

High

Low

Close