s*********o
发帖数: 14 | 1 请教各位大侠，有没有这样的time-series model:
Y_t=m(X_t, theta1)+s(X_t, theta2)*u_t
where X_t is some exogenous covariates, m and s are two functions of X_t and
/or X_t-1, X_t-2 ..., theta1 and theta2 are some parameters, u_t is error
term.
谢谢大家先 | t****g
发帖数: 715 | 2 I guess you can have whatever time series model you like. For the format you
want, why not construct a linear model as follows, where the variance of
shocks depends on regressors:
y_t= x_t*theta1 + x_t*u_t, where u_t ~ NID(0, sigma^2), var(x_t)=sigma_x^2.
Then this becomes a simple model with volaticity in shocks.
and
【在 s*********o 的大作中提到】
: 请教各位大侠，有没有这样的time-series model:
: Y_t=m(X_t, theta1)+s(X_t, theta2)*u_t
: where X_t is some exogenous covariates, m and s are two functions of X_t and
: /or X_t-1, X_t-2 ..., theta1 and theta2 are some parameters, u_t is error
: term.
: 谢谢大家先
| s*********o
发帖数: 14 | 3 Thank you for your reply. We surely can write some models satisfying such
conditions, but the problem is, are such models used in practice? | t****g
发帖数: 715 | 4 The model I just cited is used, to the best of my knowledge.
【在 s*********o 的大作中提到】
: Thank you for your reply. We surely can write some models satisfying such
: conditions, but the problem is, are such models used in practice?
| s********t
发帖数: 247 | 5 is it just classical location-scale model? It's used everywhere :P | s*********o
发帖数: 14 | 6 Yes, it's location-scale model, but for most of those models, lagged Y_t's
are also in the right-hand equation (in the location term m and/or the scale
term s). While here I want some models where the right-hand side does NOT
contain any Y's. Are models like this popular? Especially are they used in
applications? Thanks. | s*********o
发帖数: 14 | 7 The model you mentioned reminds me the normal variance-mean mixture, seems
to me it's more like a cross-section model instead of time-series, or did I
miss something?
【在 t****g 的大作中提到】
: The model I just cited is used, to the best of my knowledge.
|
|
