What inspired this Wish List Request? This is just my personal suggestion, not by a user.
Please describe the new or improved feature or solution you would like to see.
I would like to suggest to have a feature to calculate BLUP for Z*a and X*beta + Z+a.
In the current Response Screening's REML, if R-side REML is used, then the platform cannot output BLUP or conditional prediction. We cannot theoretically get unique BLUP for "a" in "y = X*beta + Z*a + e". But we can calculate unique BLUP for Z*a and X*beta + Z*a. If G = Z*Z`, then unique BLUP for Z*a is G*Inv(G + ve/vu*Identity(NRow(G))) * (y - X*b) where b is the parameter estimate for beta.
To see the current behavior, please run the script below on JMP19. There is no command to calculate BLUP or conditional prediction.
Names Default To Here(1);
Random Reset(555555);
n = 1000;
p = 20;
z = J(n, p, Random Integer(-1,1));
/* In JMP19.0, Z is always centered eveywhere. */
For(i = 1, i <= NCol(z), i++,
z[0, i] = z[0, i] - Mean(z[0, i])
);
a = J(p,1,Random Normal());
y = 3 + z*a + J(n,1,Random Normal());
/////////////////////////////////////////////////////////////////////////
dt = As Table(y||z);
Column(dt, 1)<<Set Name("y");
For(i = 2, i <= NCol(dt), i++,
Column(dt, i)<<Set Name("z"||Char(i - 1))
);
G = Z*Z`;
dt Rside = As Table(y,<<Column Names({"y"}));
dt Rside << New Column("ID", Character);
For(i = 1, i<= NRow(Z), i++,
Column(dt Rside, "ID")[i] = "ID"||Char(i);
dt Rside << New Column("ID"||Char(i), Values(G[0, i]))
);
plat Rside = dt Rside << Fit Model( Effects, Personality( "Response Screening" ), Y( :y ), ID( :ID ), Run );
What would the value be to you and/or other users if this new feature is provided? In genomic selection, we need BLUP for selectiong "better" individuals. In the current JMP, we cannot calculate BLUPs from the result by Marker Relatedness platform.
Is this an external customer request? No.
Is there a competitive software package that has this feature or solution that we should be aware of? I don't know.
Will this impact a renewal, expansion, or adoption at a customer, and does this put an account at risk if we do not address this? No
