+ Reply to Thread
Results 1 to 3 of 3

Thread: Price forecasting model

  1. #1
    Points: 163, Level: 3
    Level completed: 26%, Points required for next Level: 37

    Posts
    12
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Price forecasting model




    I am working on a problem where I have an item 'x' which is available in local market and 'x' is produced using item 'y'.Item 'y' is traded in futures market.I have historic price data available for last 2 years of both items.
    I have to forecast prices of x for next year.
    What should be my approach?

    Note:This is my first post in this forum.Pardon me if I broke any norm/rule here.

  2. #2
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Price forecasting model

    That depends entirely on how much knowledge you have of time series. There are many ways to do this and they vary from complicated to very painful.

    Essentially you are testing how one time series predicts another over time. Two obvious approaches are ARIMA with an input variable and regression with ARMA errors. Both require you to make your data stationary if they aren't already (and they almost certainly won't be stationary). Regression with ARMA errors or Regression with autoregressive error are going to be less painful to do probably.

    I have studied time series for several years, but am no expert in it. Hopefully Vinux will wander by.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  3. The Following User Says Thank You to noetsi For This Useful Post:

    TheMonkster (04-18-2015)

  4. #3
    Points: 163, Level: 3
    Level completed: 26%, Points required for next Level: 37

    Posts
    12
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Re: Price forecasting model


    Quote Originally Posted by noetsi View Post
    That depends entirely on how much knowledge you have of time series. There are many ways to do this and they vary from complicated to very painful.

    Essentially you are testing how one time series predicts another over time. Two obvious approaches are ARIMA with an input variable and regression with ARMA errors. Both require you to make your data stationary if they aren't already (and they almost certainly won't be stationary). Regression with ARMA errors or Regression with autoregressive error are going to be less painful to do probably.

    I have studied time series for several years, but am no expert in it. Hopefully Vinux will wander by.
    Thanks for response.Let me be specific here.

    I am 'Monkster Airlines'.I want to protect my business from price volatility of jet fuel cost.Jet fuel is not traded in futures market but Crude oil is traded in futures market. I have daily spot prices of jet fuel and contract prices of crude oil.
    e.g. on 19th April,spot price(Jet fuel) is 100.April contract price (crude oil) = 103,May contract price = 110,June Contract price =118 and July contract price = 109 and August price = 125. Spot and contract prices change on daily basis as per demand-supply dynamics.

    Should I find correlation between spot price on particular date and contract prices on that date for all contracts? How should I go about it?
    Please help me with approach.

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats