I am trying to fit curves to my data with the curvefit() function of the REAT package. The results look good and give me a table like this:
a b Std. Error a Std. Error b t value a t value b
Linear -1494.108840 166.1867222 523.5387976 30.50266009 -2.8538646 5.448270
Power 0.425234 2.7725836 0.4456762 0.36780235 -0.8332776 7.538243
Exponential 32.008316 0.2022666 0.4492831 0.02617634 7.7145024 7.727078
Logistic 7.076363 -0.4071710 1.6305232 0.09499830 4.3399340 -4.286087
Pr(>|t|) a Pr(>|t|) b R squared Adj. R squared F value Pr(>F)
Linear 1.275234e-02 8.582766e-05 0.6795139 0.6566221 29.68364 8.582766e-05
Power 4.186780e-01 2.716813e-06 0.8023300 0.7882107 56.82511 2.716813e-06
Exponential 2.085140e-06 2.046447e-06 0.8100606 0.7964935 59.70773 2.046447e-06
Logistic 6.790587e-04 7.535634e-04 0.5675080 0.5366157 18.37054 7.535634e-04
What I would like to do is to extract the functional equations of the curves. For linear fit, this is probably y = ax + b
. But which functions are used for the other three?
In the function code I found lm (log10(y) ~ log10(x))
for power, lm (log(y) ~ x)
for exponential and lm ( (log((y.max-y)/y)) ~ x)
for logistic.
Probably due to my faded maths skills, I don't know how to derive a function including the paramters a and b, just like the one for linear regression. Therefore, any help is appreciated.
Tried around a lot and found out this:
Linear: y = b*x + a
Power: y = a*x^b
Exponential: y = a*exp(b*x)
I am still missing the logistic function though, if someone knows it.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.