Comparing Time Series, Neural Nets and Probability Models for New Product Trial Forecasting • Eugene Brusilovskiy • Ka Lok Lee • These slides are based on the authors’ presentation at the 4 th Annual Hawaii International Conference on Statistics, Mathematics, and Related Fields
22
Embed
Comparing Time Series, Neural Nets and Probability Models for New Product Trial Forecasting Eugene Brusilovskiy Ka Lok Lee These slides are based on the.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Comparing Time Series, Neural Nets and Probability Models for New Product Trial Forecasting
• Eugene Brusilovskiy • Ka Lok Lee
• These slides are based on the authors’ presentation at the 4th Annual Hawaii International
Conference on Statistics, Mathematics, and Related Fields
2
Problem Introduction
• Goal: To predict future sales using sales information from an introductory period
• Product: A new (unnamed) soft beverage that was introduced to a test market
• Data: We have 52 weeks of sales data, which we split into training (first 39 weeks) and validation (last 13 weeks) datasets– We build the models using the training dataset and
then examine how well the models predict sales in the last 13 weeks
• The methods employed here apply to predicting the sales of any newly introduced consumer good
3
Prediction Methods Used
• Time Series– Most common technique, available in almost every
Training Data – Cumulative Sales for the First 39 Weeks
T = 39
5
Time Series
• A time-series (TS) model accounts for patterns in the past movements of a variable and uses that information to predict its future movements. In a sense a time-series model is just a sophisticated method of extrapolation (Pindyck and Rubinfeld, 1998).
6
Time Series
• Autoregressive Moving Average Model: ARMA(1,1) – generally recognized to be a good approximation for many observed time series
11 tttt yy
tt ByB 11or
7
Neural Networks
• A Neural Network (NN) is an information processing paradigm inspired by the way the brain processes information (Stergiou and Siganos, 1996).
• MLP (The Multi-Layer Perceptron) is used here
8
Neural Networks
• A Neural Network consists of neuron layers of 3 types:– Input layer– Hidden layer– Output layer
• We use two models with different MLP architectures: a model with one hidden layer and a model with a skip layer
9
Neural Networks (cont’d)
X1X2X3 X
X1X2X3 X
AND
X1X2X3 XX1X2X3 X or XX1X2X3 XX1X2X3 X or XX1X2X3 X or XX1X2X3 XX1X2X3 X or XX1X2X3 X
Given the rule on the left, we deduce the pattern on the right:
10
Neural Networks
Structure of Neural Net Models:
11
Neural Networks
• Neural Networks are especially useful for problems where– Prediction is more important than explanation– There are lots of training data– No mathematical formula that relates inputs to outputs
is known • Source: SAS Enterprise Miner Reference Help.
Neural Network Node: Reference
12
Probability Modeling
• Probability models:– Are representations of individual buying behavior – Provide structural insight into the ways in which
consumers make purchase decisions (Massy el at.,1970)• Specific assumptions of purchase process and latent
propensity (Bayesian flavor)• Explicit consideration of unobserved heterogeneity
13
Probability Modeling
• Individual purchase time or time-to-trial is modeled by “Diffusion Model”.
• Exponential-Gamma (EG), also known as the Pareto distribution (Hardie et al., 2003)
• Time to trial ~ Exponential (λ)• λ~ Gamma (r, α)
d
r
ee
rrt
1
01
14
Probability Modeling
• After solving the integral, the cumulative probability function becomes:
• F(t) =
• LL =
• Estimation uses Excel Solver
TF
tFtFSales
T
tt
1ln
1
r
t
1
15
16
Results
Exp. Gamma
Neural Nets
Time Series
Mean Absolute Percentage Error (MAPE)
2.7% 9.0% 5.5%
T
SalesActual
SalesredictedPSalesActual
MAPE
T
t t
tt
1
Where T is the total number of time periods (weeks). Here, t=1 is the first validation week (week 40)
• All three models do a relatively good job predicting future sales, but Exponential Gamma is the best
• Include covariates• Different training periods• Perform comparative analysis for other areas of
forecasting – Customer Lifetime Value
22
References
• Hardie B. G.S., Zeithammer R., and Fader P. (2003), Forecasting New Product Trial in a Controlled Test Market Environment, Journal of Forecasting, 22: 391-410
• Massy, W.F., Montgomery, D.B. and Morrison, D.G. (1970), Stochastic Models of Buying Behavior, The M.I.T. Press, 464 pp.
• Pindyck, R.S. and Rubinfeld D.L. (1998), Econometric Models and Economic Forecasts, Irwin/McGraw-Hill.
• Stergiou, C., & Siganos, D. (1996), Introduction to Neural Networks. Available online at www.doc.ic.ac.uk/~nd/surprise_96/journal/vol4/cs11/report.html