14.09.2015 Views

Matvec Users’ Guide

Matvec Users' Guide

Matvec Users' Guide

SHOW MORE
SHOW LESS
  • No tags were found...

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

11.1. GENERALIZED LINEAR MODEL 73<br />

11.1.4 Threshold Model<br />

<strong>Matvec</strong> can also be used to fit threshold models based on either an underlying normal, Pr(Z ≤ x) = Φ(x),<br />

or logistic, Pr(Z ≤ x) = e x /(1 + e x ) distribution. Suppose data has been collected on calving difficulty to<br />

examine the effect of gender of the calf on calving difficulty. The data are given in Table 11.3.<br />

Table 11.3: Calving difficulty data.<br />

Calving Difficulty<br />

Sex Easy Moderate Difficult<br />

Male 169 16 15<br />

Female 161 23 16<br />

The threshold link function in <strong>Matvec</strong> requires that ordinal data with c categories be coded as 0, 1, . . . ,<br />

c − 1. The probability that an observation falls in category i is given by<br />

Pr(Y = i) = Pr(τ i−i + η < Z ≤ τ i + η)<br />

where τ −1 = −∞, τ 0 = 0 and τ c = ∞. The data for this study are stored in the file calve.dat:<br />

Male 0 169<br />

Male 1 16<br />

Male 2 15<br />

Female 0 161<br />

Female 1 23<br />

Female 2 16<br />

and are read using:<br />

D=Data();<br />

D.input("calve.dat","Sex $ Score n");<br />

For this example, the linear predictor for observation j will include an intercept and sex main effect. The<br />

threshold between moderate (1) and difficult (2) will only include an intercept. The model is specified by:<br />

M=Model();<br />

M.equation("Score= intercept Sex,Score=intercept");<br />

M.link("thresh",0)<br />

M.param(1);<br />

M.weight("n");<br />

M.fitdata(D);<br />

The M.param(1) indicates that a probit threshold model will be used. Alternatively M.param(0) can be<br />

used to fit a logistic threshold model. The analysis is completed using:<br />

M.glim(20)<br />

M.save("calve.out");<br />

M.contrast("Sex",[1 0 ,-1 0])<br />

Resulting in the following:<br />

Threshold Link Function<br />

Parameters: Probit/Logit (scalar)) 1=probit 0=logit<br />

Terms: Eta1, Thresh1, Thresh2, ...<br />

y 0..numtraits<br />

0<br />

0.48819

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!