Comments (3)
The standard errors are for each model parameter, and represents the standard deviation if you assume that the model is the correct form for each of the model parameters (that assumption includes that the break point locations are correct).
I still need to investigate an approach of a piecewise linear standard deviation... but in the meantime you can calculated the unbiased prediction variance and use it to generate your 95% confidence interval of your model.
I include a function to calculate the sum of squares of the residuals, which is the hard part.
Then all you need to do is calculate the unbiased prediction variance. It's usually easier though to work with standard deviations instead.
# calculate the sum of the square of the residuals
ssr = my_pwlf.fit_with_breaks(my_pwlf.fit_breaks)
# calculate the unbiased standard deviation
sigma = np.sqrt(ssr / (my_pwlf.n_data - my_pwlf.n_parameters))
# sigma can be used as a prediction variance
# plot the results
plt.figure()
plt.plot(x, y, 'o')
plt.plot(xHat, yHat, '-')
plt.fill_between(xHat, yHat - (sigma*1.96), yHat + (sigma*1.96), alpha=0.1,
color="r")
plt.show()
from piecewise_linear_fit_py.
So I've finally implemented a prediction variance that represents the uncertainty due to the lack of data. My formula is based on one from http://www2.mae.ufl.edu/haftka/vvuq/lectures/Regression-accuracy.pptx Prediction variance is the squared version of standard error. Standard error should not be confused with standard errors.
An example on how to use the prediction variance is found in https://github.com/cjekel/piecewise_linear_fit_py/blob/master/examples/prediction_variance.py
I'll upload a new version to pypi shortly. I believe this change is what you were looking for.
from piecewise_linear_fit_py.
Thank you very much :)
from piecewise_linear_fit_py.
Related Issues (20)
- .fit() fails with 1 segment HOT 5
- How to force the fit process to have a fixed Intercept? HOT 1
- Can I get y_values if I have only x and slopes values? HOT 1
- Limit the slope of each segment of the curve HOT 1
- Re-constructing Piecewise PWLF HOT 2
- Set Slope of Segment to 0 HOT 2
- How to fit multiple functions simultaneously HOT 3
- Hi, i want to make sure that there are no fitted lines between points that are too far apart i.e. set a min value( fragment optimization) how to achieve this? HOT 1
- Why last beta is always positive? HOT 3
- How to prevent poor fitting HOT 4
- Forcing the segments to be integers and not float
- p values does not seem accurate HOT 3
- Error for coefficients of linear equations HOT 7
- pwlf with unknown line segments HOT 9
- assure the slopes to be lower and lower HOT 1
- divide by zero error in calc slopes if two break points are the same, or if a breakpoint is on the boundary HOT 2
- Issue using .fit() HOT 5
- How to plot segments with fit_breaks information HOT 4
- support random seed on init
- How to calculate prediction intervals? HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from piecewise_linear_fit_py.