Group Abstract Group Abstract

Message Boards Message Boards

0
|
5.5K Views
|
3 Replies
|
0 Total Likes
View groups...
Share
Share this post:
GROUPS:

Create a Correlation Table with financial data

Posted 12 years ago

Hello,

I am just discovering wolfram alpha and i do not know how to programme so this will be basic but I have not seen the question so I decided to ask it myself. I think that the creation table could be useful for finance. There is an exemple of a correlation table with wolfram programming: http://reference.wolfram.com/language/example/CreateACorrelationTable.html.

  1. Firstly, there is a need to choose financial members usch as the members of a stock index (nasdaq,...). This could be similar to the exemple (Get the members of the G8: In Click for copyable input Out[1]= ) but with the companies of a financial index. Is it possible to select the members of the DJIA or the NASAD?

  2. Then the second step is to play with the characteritics of these members (capitalisation, PERRatio,...).

Has someone tried this already? Otherwise, is there a way to produce an automatical financial analysis of a stock with wolfram alpha?

Thanks. I hope that it will also be useful to others people.

POSTED BY: Christophe petit
3 Replies
Posted 22 days ago

Hi Christophe,

Good question — this is actually a very common use case for Wolfram Language in finance.

Yes, it is possible to work with members of major indices like DJIA or NASDAQ, but this generally requires using Wolfram Language (Mathematica) rather than plain Wolfram|Alpha queries. Wolfram has built-in financial data functions that make this much easier.

For example, index constituents can be retrieved via entities, such as FinancialIndex and FinancialInstrument. Once you have the list of tickers, you can pull time series data (prices, returns) and then compute correlations using built-in functions like Correlation or CorrelationMatrix.

Regarding fundamentals (market cap, P/E ratio, etc.), Wolfram provides access to many of these properties through entity properties, although coverage can vary by market and by company.

As for automatic financial analysis: Wolfram|Alpha itself is more query-based and limited, but in Wolfram Language you can absolutely automate analysis — from correlation tables to factor comparisons and basic portfolio analytics — provided the data is available.

So in short:

Yes, index members can be selected programmatically

Yes, correlation tables are very feasible

Full automation is much more realistic in Mathematica than in Wolfram|Alpha alone

Hope this helps, and curious to hear if others here are using Wolfram for portfolio or factor analysis.

POSTED BY: Jasmin Best

Well I thoght that I was using wolfram/alpha but I must have switched to mathematica. Thank you for your answer regarding the index. I thought that it would be interesting to find covariances and correlations between characteristics of stocks ( sector, PERatio, margins,...). I was wondering if there is a embryo of big data ai or of artificial intelligence such as grok that looks for patterns in datas. But I admit that I am a tourist and I do not know much. Because from my understanding, the end goal of all these wolfram products is to find patterns in datas and play in an intuitive way with all the patterns that could be useful for business and for science. For exemple in finance, if i ask to look for correlations for the NASDAQ between 1990 and 2000 then it would find a correlation of 0.99 between energy stocks and inflation,... In science, it could find newton's law because all objects that fall have an acceleration of a. Therefore you need to record as much data as possible and then to find patterns in all the datas. Obviously this does not exist yet but as I am just discovering the wolfram products and having read on wikipedia that stephen wolfram wrote " a new kind of science", I was wondering if the embryo of such a general pattern recognition exists so that you only have to give it some data input for patterns outputs. If there is a section that deals more specifically with this, pelase send me the link. Otherwise, thank you for the nasdaq tip, I will use it and play with it. Thank you again, that is helpful.

POSTED BY: Christophe petit

Are you using Wolfram|Alpha or Mathematica? I can't tell.

The example at reference.wolfram.com is for Mathematica. You can't exactly program with Wolfram|Alpha.

  1. There are a number of different ways to get the members of an index. FinancialData can pull this information from Yahoo Finance in many cases. For example the members of NASAQ can be obtained with FinancialData["NASDAQ", "Members"]. Sometimes you can get this list from a wolframalpha query.
POSTED BY: Sean Clarke
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard