Data for Stock Market Prediction

Answers

• Options

  ronmac Member Posts: 11 Contributor II

  August 2011

  I like to start my data analysis process by collecting various financial data series and run the RM Mutual Information Operator on the series. That helps gets me in the right "stadium" with the data sets to use The Correlation Operator is useful also. Then I work up to getting on the "playing filed" using other RM operators. Lately I have been using the Kmeans Cluster Operator to help find some useful predictive patterns. Yahoo Pipes has some great tools for data collecting.http://pipes.yahoo.com/pipes/ Here is an Excel Macro I use to pull Yahoo data directly into Exce. Then I read the workbook into RM. Makes it fairly easy and fast to find useful relationships in your data. Sorry there is no short cut and I don't think anybody that believes they have discovered the "Holy Grail" is going to come forward with the secret recipe
  A very good video I just watched covers this “Markets are Efficient if and only if P = NP” http://www.youtube.com/watch?v=7iOJZZFDKpc&feature=player_embedded
  
  Cheers.
  Ron McEwan
  

  Sub GetData()
  
      Dim QuerySheet As Worksheet
      Dim DataSheet As Worksheet
      Dim EndDate As Date
      Dim StartDate As Date
      Dim Symbol As String
      Dim qurl As String
      
      Application.ScreenUpdating = False
      Application.DisplayAlerts = False
      Application.Calculation = xlCalculationManual
      
      Set DataSheet = ActiveSheet
    
          StartDate = DataSheet.Range("B2").Value
          EndDate = DataSheet.Range("B3").Value
          Symbol = DataSheet.Range("B4").Value
          Range("C7").CurrentRegion.ClearContents
          
  'construct the URL for the query
          
          qurl = "http://chart.yahoo.com/table.csv?s=" & Symbol
          qurl = qurl & "&a=" & Month(StartDate) - 1 & "&b=" & Day(StartDate) & _
              "&c=" & Year(StartDate) & "&d=" & Month(EndDate) - 1 & "&e=" & _
              Day(EndDate) & "&f=" & Year(EndDate) & "&g=" & Range("C3") & "&q=q&y=0&z=" & _
              Symbol & "&x=.csv"
          Range("c5") = qurl
                     
  QueryQuote:
               With ActiveSheet.QueryTables.Add(Connection:="URL;" & qurl, Destination:=DataSheet.Range("C7"))
                  .BackgroundQuery = True
                  .TablesOnlyFromHTML = False
                  .Refresh BackgroundQuery:=False
                  .SaveData = True
              End With
              
              Range("C7").CurrentRegion.TextToColumns Destination:=Range("C7"), DataType:=xlDelimited, _
                  TextQualifier:=xlDoubleQuote, ConsecutiveDelimiter:=False, Tab:=True, _
                  Semicolon:=False, Comma:=True, Space:=False, other:=False
              
              Range(Range("C7"), Range("C7").End(xlDown)).NumberFormat = "mmm d/yy"
              Range(Range("D7"), Range("G7").End(xlDown)).NumberFormat = "0.00"
              Range(Range("H7"), Range("H7").End(xlDown)).NumberFormat = "0,000"
  UpdateScale
  RemoveNames
  
  
  'turn calculation back on
      Application.Calculation = xlCalculationAutomatic
      Application.DisplayAlerts = True
      Range("C7:H2000").Select
      Selection.Sort Key1:=Range("C8"), Order1:=xlAscending, Header:=xlGuess, _
          OrderCustom:=1, MatchCase:=False, Orientation:=xlTopToBottom
      Range("B4").Select
  Paste
  
  End Sub

  0
• Options

  wessel Member Posts: 537 Maven

  August 2011

  Hey,
  
  I'm not looking for some holy grail.
  
  When searching ScienceDirect there are lots of papers that analyze stock market data,
  using many more variables then just past prices, like:
  Dividends for the index, bid-ask spread of T-bills, T-bill holding period, nominal stock returns of the index, excess returns of the index, various duration treasury rates, various duration CD rates, corporate bond yields, the producer price index, the consumer price index, the industrial production index, M1 money supply, various treasury term spreads, and various default spreads between BAA bonds and various treasuries.
  
  I want to do similar research, but I have no idea where to download such data.
  Looked at yahoo pipes, although it is cool, I don't think it helps with this problem.
  
  Best regards,
  
  Wessel
  

  0
• Options

  ronmac Member Posts: 11 Contributor II

  August 2011

  Some of the data you are looking for can be found at the St Louis Federal Reserve site http://research.stlouisfed.org/fred2/. They have a free Excel tool to simplify data retrieval http://research.stlouisfed.org/fred-addin/

  0