Commit 0c6a8e90 authored by pfreon's avatar pfreon
Browse files

Minor changes on the model text files used to describe them when a given model...

Minor changes on the model text files used to describe them when a given model is declined with of without a constant term (intercept).

When a given model is declined with of without a constant term (intercept) the user is informed and suggestion to use one or the other model is provided.
parent 3ed0e774
......@@ -6,9 +6,15 @@ linear relationship between the CPUE and the environmental variable V.
This latter variable is supposed to only influence the abundance of the
stock (surplus production) and not the catchability coefficient.
The particularity of this model is the absence of constant term in its
equation. After fitting this model, if the fit is not good it is suggested
to try the following model which has a constant term but fewer degrees of
freedom: CPUE = a + b V + c E
The particularity of this model is the absence of an intercept parameter
(constant term a) in its equation. After fitting this model, if the fit is
not good it is suggested to try fitting and validating the following related
model which includes an intercept parameter but, as a result, has fewer degrees
of freedom:
CPUE = a + b V + c E.
For doing this, please use click on "Modelization" and then on
"Fit a model directly" in the main menu of CLIMPROD.

\ No newline at end of file
......@@ -7,9 +7,14 @@ environmental variable V. This latter variable is supposed to only influence
the abundance of the stock (surplus production) and not the catchability
coefficient.
After fitting this model, if the value of the parameter a is not
significantly different from zero, it is suggested to try the following
model which has no constant term but more degrees of freedom:
CPUE = a V exp(b E).
After fitting this model, if the value of the parameter a is not significantly
different from zero, it is suggested to try fitting and validating the following
model which has no intercept parameter (constant term a) but, as a result, has
more degrees of freedom:
CPUE = a V exp(b E).
For doing this, please use click on "Modelization" and then on
"Fit a model directly" in the main menu of CLIMPROD.

\ No newline at end of file
......@@ -7,12 +7,15 @@ and the environmental variable V. This last variable is supposed to
only influence the abundance of the stock (surplus production) and not
the catchability coefficient.
The particularity of this model is the absence of constant term in its
equation. After fitting this model, if the fit is not good it is suggested
to try the following model which has a constant term but fewer degrees of
freedom:
The particularity of this model is the absence of intercept parameter
(constant term) in its equation. After fitting this model, if the fit is
not good it is suggested to try fitting and validating the following
model which includes an intercept term (a) but, as a result, has fewer
degrees of freedom:
CPUE = (a + b V) exp(c E).
For doing this, please use click on "Modelization" and then on
"Fit a model directly" in the main menu of CLIMPROD.
CPUE = (a + b V) exp(c E).

\ No newline at end of file
......@@ -9,8 +9,14 @@ those two variables. In case of exceptionally good correlation
historical data-series, but not for prediction beyond the limits
of observed values of fishing effort and 4nvironmental variations.
It uses only two parameters and therefore presents more degrees of
freedom than most of the other available models. If the fit is not
good, it is suggested to fit the same model with a constant term,
which leads to a three parameter equation: CPUE = a + b V^c
The particularity of this model is the absence of intercept parameter
(constant term) in its equation. After fitting this model, if the fit is
not good it is suggested to try fitting and validating the following
model which includes an intercept term (a) but, as a result, has fewer
degrees of freedom:
CPUE = a + b V^c
For doing this, please use click on "Modelization" and then on
"Fit a model directly" in the main menu of CLIMPROD.

\ No newline at end of file
......@@ -9,8 +9,13 @@ those two variables. In case of exceptionally good correlation
historical data-series, but not for prediction beyond the limits
of observed values of fishing effort and environmental variations.
It is a flexible equation with three parameters. If, after fitting
the value of parameter, a is not significantly different from zero,
it could be interesting to fit a model which used only two parameters
and no constant term: CPUE = a V^b.
After fitting this model, if the value of the parameter a is not significantly
different from zero, it is suggested to try fitting and validating the following
model which has no intercept parameter (constant term a) but, as a result, has
more degrees of freedom:
CPUE = a V^b.
For doing this, please use click on "Modelization" and then on
"Fit a model directly" in the main menu of CLIMPROD.

\ No newline at end of file
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment