Ten months ago I made a movie demonstrating a simple time series conversational engine. (I opened my "Quantile regression" presentation at WTC 2014 with that movie.)
Today I published a blog post that discusses the design and programming of the conversational engine: Simple time series conversational engine .
The conversational engine responds to commands for:
- loading time series data (weather data, stock data, or data files),
- finding outliers,
- analysis by curve fitting,
- plotting, and
- help and state changes.
A package with all of the code of the conversational engine can be downloaded from GitHub, [12] and executed with Import:
Import["https://raw.githubusercontent.com/antononcube/MathematicaForPrediction/master/Examples/SimpleTimeSeriesConversationalEngine.m"]
More about functional parsers, which were essential for the building of the conversational engine, can be read in "Functional parsers" by Fokker.
Here are some commands to try:
- load temperature for Atlanta
- load stock price of NYSE:AAPL
- find regression quantile for temperature of Atlanta
- find outliers for temperature of Atlanta
- least squares fit with Join[{1,x},Table[Sin[i*x/10^6.],{i,0,10}]]
