Thanks for your response Gustav,
The dots in the models mean that there are other effects in the models.
In relation to the model, it would be like this:
Q(i) = Q(i-1) + b*max(s(i) - s(0), 0) + c*min(s(i) - s(0), 0) If( s(i) >
s(0)) then
max(s(i) - s(0), 0) = s(i)-s(0)
min(s(i) - s(0), 0) = 0
There is an accumulative positive effect and Q(1) is greater than Q(0)
If( s(1) < s(0)) then
max(s(i) - s(0), 0) = 0
min(s(i) - s(0), 0) = s(i)-s(0)
There is an accumulative negative effect and Q(1) is lower than Q(0)
For i=1:
Q(1) = Q(0) + b*max(s(1) - s(0), 0) + c*min(s(1) - s(0), 0) If( s(1) > s(0))
then
max(s(1) - s(0), 0) = s(1)-s(0)
min(s(1) - s(0), 0) = 0
If( s(1) < s(0)) then
max(s(1) - s(0), 0) = 0
min(s(1) - s(0), 0) = s(1)-s(0)
In the model the value of Q depends on the value of s. If that value is
above a threshold, Q increases and if the value of s is below the threshold
Q decreases. Then, when Q is above a threshold (different that the threshold
of s) it has an effect.
In my particular case, s is temperature and Q would be the amount of
temperature (or something similar) that is accumulated.
-----Original Message-----
From: Gustaf Rydevik [mailto:gustaf.rydevik@gmail.com]
Sent: Friday, September 28, 2007 11:13 AM
To: Manuel Ramon
Cc: r-help@r-project.org
Subject: Re: [R] Is there a model like that in R?
On 9/28/07, Manuel Ramon <mramon@uga.edu> wrote:
> Hi to everyone,
>
>
>
> I am starting to work with a model that is not familiar to me.
>
> The model would be like that:
>
>
>
> y = . - a*max(Q(i) - Q(0), 0) + .
>
>
>
> where Q(i) is the accumulated effect of a variable at time i and Q(0)
> a threshold above it there is effect on y. The value of Q(i) could be
> estimated as:
>
>
>
> Q(i+1) = Q(i) + b*max(s(i) - s(0), 0) + c*min(s(i) - s(0), 0) + .
>
>
>
> Where s would be the effect that produces the accumulate effect of Q,
> s(i) would be this effect at time i and s(0) another threshold above
> it the accumulative effect is produced. The coefficient b would be the
> rate of accumulation and c the rate of decay.
>
>
>
> What kind of model is it? Is it somewhat similar to time series?
>
>
>
> I appreciate your help.
>
>
>
> Manuel Ramon
>
I'm not sure what you're describing, but something is wrong for your
definition of Q(i).
For i=1:
Q(1) = Q(0) + b*max(s(0) - s(0), 0) + c*min(s(0) - s(0), 0) +
.=Q(0)+b*0+c*0=Q(0).
Thus, Q(n)=Q(n-1)=...=Q(0)
In addition, what is intended by the dots and the final plus in "y = .
- a*max(Q(i) - Q(0), 0) + ."?
I think you have to describe things a bit more careful, if we are to
understand what's happening.
Best,
Gustaf
--
Gustaf Rydevik, M.Sci.
tel: +46(0)703 051 451
address:Essingetorget 40,112 66 Stockholm, SE skype:gustaf_rydevik
Manuel
[[alternative HTML version deleted]]