>I'm fairly new to SQL server 2005
database. The client wants to use weighted parameters(5) or areas
of a project, rated from 1-5 on a scale of 1 being the highest, and
the Cost of each project.
i.e. Project A Param1=1, Param2=4 , Param3=1, Param4=2, Param5=5 , Cost= $50000
>Using the least squares fit they want to be able to calculate a
>projected cost for a project with same weighted parameters.
With this being said I was wondering what I could use to do this using the information from the SQL Server. >Example:
of a project, rated from 1-5 on a scale of 1 being the highest, and
the Cost of each project.
i.e. Project A Param1=1, Param2=4 , Param3=1, Param4=2, Param5=5 , Cost= $50000
>Using the least squares fit they want to be able to calculate a
>projected cost for a project with same weighted parameters.
With this being said I was wondering what I could use to do this using the information from the SQL Server. >Example:
Currently this is a formula used for calculating the least square's fit using a matrixs
Each Row of completed Project Cost for a completed project
ex. Matrix A = 1, u1, v1, w1, x1 Matrix b = M1
1, u2, v2, w2, x2 M2
1, u3, v3, w3, x3 M3
1, un, vn, wn, xn Mn
Each Row of completed Project Cost for a completed project
ex. Matrix A = 1, u1, v1, w1, x1 Matrix b = M1
1, u2, v2, w2, x2 M2
1, u3, v3, w3, x3 M3
1, un, vn, wn, xn Mn
where u,v,x are weighted parameters where M is a a project
x=(A^T * A)^-1 * (A^T * b) x is the function that defines the least
squares curve for the data set
squares curve for the data set
Comment