Gnuplot

Linear fit (Regression) with gnuplot

> f(x) = m*x + b
> fit f(x) 'plotdata1_00000.txt' using 1:2 via m,b
 
 Iteration 0
 WSSR        : 20032.9           delta(WSSR)/WSSR   : 0
 delta(WSSR) : 0                 limit for stopping : 1e-005
 lambda   : 0.814969
 
initial set of free parameter values
 
m               = 1
b               = 1
/
 
 Iteration 1
 WSSR        : 955.717           delta(WSSR)/WSSR   : -19.9611
 delta(WSSR) : -19077.2          limit for stopping : 1e-005
 lambda   : 0.0814969
 
resultant parameter values
 
m               = -17.086
b               = -2.65671
/
 
 Iteration 2
 WSSR        : 941.634           delta(WSSR)/WSSR   : -0.0149562
 delta(WSSR) : -14.0832          limit for stopping : 1e-005
 lambda   : 0.00814969
 
resultant parameter values
 
m               = -18.382
b               = -2.03949
/
 
 Iteration 3
 WSSR        : 941.634           delta(WSSR)/WSSR   : -1.45998e-008
 delta(WSSR) : -1.37477e-005     limit for stopping : 1e-005
 lambda   : 0.000814969
 
resultant parameter values
 
m               = -18.3833
b               = -2.03882
 
After 3 iterations the fit converged.
final sum of squares of residuals : 941.634
rel. change during last iteration : -1.45998e-008
 
degrees of freedom    (FIT_NDF)                        : 98
rms of residuals      (FIT_STDFIT) = sqrt(WSSR/ndf)    : 3.09976
variance of residuals (reduced chisquare) = WSSR/ndf   : 9.60851
 
Final set of parameters            Asymptotic Standard Error
=======================            ==========================
 
m               = -18.3833         +/- 1.074        (5.841%)
b               = -2.03882         +/- 0.6153       (30.18%)
 
 
correlation matrix of the fit parameters:
 
               m      b
m               1.000
b              -0.864  1.000    
 
> plot 'plotdata1_00000.txt', f(x)

Exponential fit (regression) using gnuplot

> f(x) = A*exp(B*x)
> fit f(x) 'plotdata1_00000.txt' using 1:2 via A,B
...
 
After 9 iterations the fit converged.
final sum of squares of residuals : 498.221
rel. change during last iteration : -6.28782e-008
 
degrees of freedom    (FIT_NDF)                        : 98
rms of residuals      (FIT_STDFIT) = sqrt(WSSR/ndf)    : 2.25475
variance of residuals (reduced chisquare) = WSSR/ndf   : 5.08389
 
Final set of parameters            Asymptotic Standard Error
=======================            ==========================
 
A               = -3.68297         +/- 0.2443       (6.634%)
B               = 1.91544          +/- 0.08438      (4.405%)
 
 
correlation matrix of the fit parameters:
 
               A      B
A               1.000
B               0.962  1.000 
 
> plot 'plotdata1_00000.txt', f(x)

How to do colored scatter plots?

See here:

reset
set xrange [-2:2]
set yrange [-2:2]
set pm3d
splot exp(-x*x)*exp(-y*y)

And how do I get gray-scaled scatter plots?

reset
set xrange [-2:2]
set yrange [-2:2]
set pm3d
set palette gray
splot exp(-x*x)*exp(-y*y)

How to show colored contours?

reset
set xrange [-2:2]
set yrange [-2:2]
set pm3d at b
splot exp(-x*x)*exp(-y*y)

How to generate a histogram with gnuplot?

bin_width = 0.01
bin_number(x) = floor(x/bin_width)
print bin_number(0.05)
# will show '5'
rounded(x) = bin_width * (bin_number(x) + 0.5)
plot 'mydata.txt' u (rounded($1)):(1.0) t 'data' smooth frequency w boxes
 
public/gnuplot.txt · Last modified: 2013/03/06 18:30 (external edit) · []
Recent changes RSS feed Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki