trendinBot intends to be a tool helping the human analyst (financial advisor, pollist, simulation engineer, etc.) to manage apparently unrelated information
hence, and excepting for the case of very simple situations, the actual version of the program should not be used as a standalone predictor
skip to main |
skip to sidebar
go numerics in the easiest way
our principles
- ACCURACY
- simpler = better
- if possible, nicer
our policy
- combinatorics better than simplifying statistics
contents
- ___
- 1- trad. statistics
- > intro
- > trend finding
- - a] regressions
- + b] forecasting
- ----- time series
- ----- extrapolations
- ----- PLS regression
- ----- p. components
- ----- (probability)
- ___
- 2- new theory
- ___
- 3- case studies
- > beauty contest
- > clothes shop
- > anatomical analysis
- > new product release
- ___
- _________
trendingBot
- 1. feed it -raw data-
- 2. let it run
- 3. equation or "trend not found"
NEW IDEAS
- any situation = multiple behaviours
- any behaviour = mathematically describable
- error in predictions = wrong variable selection
- user intervention = wrong model delimitation
- _________
trad. statistics
- lots of user-defined parameters
- lots of simplifications
- lots of assumptions
- complex methods
- not to-the-point solutions
- "cannot" has no meaning
doe
- same as statistics
- max/min = too simplified
- determine how measurements have to be performed