Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Approximate cost calculation

  1. Jul 16, 2008 #1
    I have the following formula -
    a*k1 + b*k2 + c*k3 = P
    This is a formula i'm trying to build. It will be used to calculate price of a specific good (P) according to criterions k1,k2,k3. Every k have its own value, thats why its being multiplied by coefficients a,b and c.
    I have a history of many previous purchases where k1,k2,k3 and P are known and i need to approximately calculate a,b and c according to this.

    Hope i explained this well.
    Can someone help me with this?
  2. jcsd
  3. Jul 16, 2008 #2
    Just an idea:

    [tex]P[/tex] is a function of [tex]k_1, k_2,[/tex] and [tex]k_3.[/tex]
    So the multivariable formula is in the form [tex]P(k_1, k_2, k_3) = ak_1 + bk_2 + ck_3[/tex].

    If you have many values of [tex]P[/tex] and [tex]k[/tex] in the formula [tex]P(k) = ak[/tex], you would run a single variable regression on the data to find a reasonable value of [tex]a[/tex]. Thus far, I am not fluent in multivariable regression, so here is a http://www.nd.com/NSBook/NEURAL%20AND%20ADAPTIVE%20SYSTEMS17_Regression_for_Multiple_Varia.html" [Broken] that might help.
    Last edited by a moderator: May 3, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook