# [fpr 392] Dr. Browne's Special Seminar at ISM

\$BA0EDCiI'(B

```                                                   96.9.17
\$B4X783F0L(B
\$B!!!!!!(B    \$B!!%V%i%&%s@h@8FCJL%;%_%J!<\$N\$*CN\$i\$;(B

\$B2<5-\$NDL\$jFCJL%;%_%J!<\$r3+:E\$7\$^\$9!#;22CL5NA\$G\$9!#J3\$C\$F\$4(B
\$B;22C2<\$5\$\$!#\$J\$*!"K\%;%_%J!<\$OE}7W?tM}8&5f=j\$NFCJL%;%_%J!<(B
\$B\$H!"Bg3XF~;n%;%s%?!l!!=j!!!'E}7W?tM}8&5f=j%;%_%J!<<e\$X!K(B
URL: http://www.ism.ac.jp  \$B\$KCO?^\$"\$j!#(B
\$B!!MW!!;]!!!'!J\$d\$d>\:Y\$J\$b\$N\$rBW\$-!"99\$KC;\$/\$5\$;\$FBW\$-\$^\$7\$?!K(B
Latent curve analysis is intended for situations where obser-
vations are taken over time on a learning task at equally
spaced intervals for each of a number of sujects.  This yields
a N*p_{T} data matrix where N represents the number of subjects
and p_{T} the number of trials per subject.  Each subject's
vector of obervations is regarded as the weighted sum of m_{T}
fixed vectors, each representing p_{T} equally spaced readings
on a basis function, plus a random vector of errors.  The model
is closely related to the factor analysis model.  An example is
presented where the model is fitted to data by maximum likeli-
hood.

\$BLd\$\$9g\$o\$;!'2<5-\$^\$G(B
---------------------------------------------
\$BA0ED!!CiI'!wE}7W?tM}8&5f=j(B
maeda (at) ism.ac.jp
(TEL: 03-5421-8734, FAX: 03-5421-8796)
---------------------------------------------

```

\$B%9%l%C%II=<((B \$BCx \$BF|IU=gI=<((B \$B%H%C%W%Z!<%8(B

\$B\$3\$3\$O(B\$B?4M}3X8&5f\$N4pAC%a!<%j%s%0%j%9%H(B\$B\$KEj9F\$5\$l\$?2a5n\$N5-;v\$r7G:\\$7\$F\$\$\$k%Z!<%8\$G\$9!#(B