Comments (3)
Hi @rustyconover, currently there is no trick that you can do to speed that up. But we are working on a branch refactoring-to-0.4
that will give you some options that will be merged soon. Some of the options are:
- Model the seasonality as a cycle, this reduces the number of states.
- Implement a custom Kalman Filter for the structural model equation.
As you grow the size of the system matrices, i.e. T
the longer it takes to estimate because the Kalman Filter equations will cost more.
from statespacemodels.jl.
@guilhermebodin Thank you for your reply. I understand now why having long cycles can be suboptimal. Maybe I will downsample the data such that it results in data points.
Just in case you're interested, using your module has led me to the top of the real-time prediction leaderboard at microprediction.org.
from statespacemodels.jl.
@rustyconover I would love know more about the particular application. The refactoring-0.4 will handle real-time predictions in a much more efficient way.
from statespacemodels.jl.
Related Issues (20)
- Scenarios of SARIMA
- ExponentialSmoothing: AssertionError ub > lb HOT 4
- Add keyword arg to ignore the hessian calculation HOT 1
- get_smoothed_state performance issues for larger time series HOT 5
- LAPACKException for missing/NaN values HOT 3
- Allow Polynomials v2 in [compat] HOT 1
- Possible Issue with Std.Error in SARIMA() HOT 2
- BasicStructuralExplanatory works ONLY for s = 12. HOT 3
- simulate_scenarios for BasicStructuralExplanatory fails if seasonality is not 12 HOT 2
- Feature suggestion: state space models for intermittent demand
- Example in the Quick Start not working HOT 1
- Getting DomainError when fitting some models HOT 2
- max_q must be strictly positive? (auto arima) HOT 3
- Unobserved components model with smooth trend is broken
- Enhancement Request - Add ability to specify innovations for simulation HOT 2
- Linear univariate time variant model implemented? HOT 8
- Holt-Winter Model HOT 1
- Can not fit MA(q) HOT 10
- Cannot predict for more steps than number of observations
- Fix all parameters and calculate just the likelihood for a filter 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 statespacemodels.jl.