480 likes | 497 Views
This paper introduces actuaries to neural networks, showing their similarities to conventional statistics and discussing where their use can be helpful. It elucidates how to interpret neural network models and their advantages over other data mining methods. The paper covers supervised and unsupervised learning in neural networks, along with examples of fitting curves, approximating functions, and handling correlated variables. It also explores interpreting neural networks, interactions in modeling, and visualizing neural network results.
E N D
Neural Networks Demystifiedby Louise FrancisFrancis Analytics and Actuarial Data Mining, Inc.louise_francis@msn.com
Objectives of Paper • Introduce actuaries to neural networks • Show that neural networks are a lot like some conventional statistics • Indicate where use of neural networks might be helpful • Show how to interpret neural network models
Data Mining • Neural networks are one of a number of data mining techniques • Methods primarily developed in artificial intelligence and statistical disciplines to find patterns in data • Typically applied to large databases with complex relationships
Some Other Data Mining Methods • Decision trees • Clustering • Regression splines • Association rules
Some Data Mining Advantages • Nonlinear relationships • Interactions • Multicollinearity
Data Mining: Neural Networks • One of more established approaches • Somewhat glamorous • AI description: they function like neurons in the brain
Neural Networks: Disadvantages • They are a black box • User gets a prediction from them, but the form of the fitted function is not revealed • Don’t know which variables are the most important in the prediction
Kinds of Neural Networks • Supervised learning • Multilayer perceptron • Also known as backpropagation neural network • Paper explains this kind of NN • Unsupervised learning • Kohonen neural networks
THREE LAYER NEURALNETWORK Hidden Layer (Processes Data) Input Layer (Input Data) Output Layer (Predicted Value) The MLP Neural Network
The Activation Function • The sigmoid logistic function
Other • Data is usually normalized • Usually both independent and dependent variables transformed to lie in range between 0 and 1
Fitting the curve • Typically use a procedure which is like gradient descent
Table 4 W0 W1 Node 1 -4.107 7.986 Node 2 6.549 -7.989 Fitted Weights
Table 5 W0 W1 W2 6.154 -3.0501 -6.427 Hidden Layer
Table 6 Computation of Predicted Values for Selected Values of X (1) (2) (3) (4) (5) (6) (7) ((1)-508)/4994 6.15-3.05*(3)-6.43*(4) 1/(1+exp(-(5)) 6.52+3.56*(6) X Normalized X Output of Node 1 Output of Node 2 Weighted Hidden Node Output Output Node Logistic Function Predicted Y 508.48 0.00 0.016 0.999 -0.323 0.420 7.889 1,503.00 0.22 0.088 0.992 -0.498 0.378 7.752 3,013.40 0.56 0.596 0.890 -1.392 0.199 7.169 4,994.80 1.00 0.980 0.190 1.937 0.874 9.369 Selected Fitted Values for function
Universal Function Approximator • The multilayer perceptron neural network with one hidden layer is a universal function approximator • Theoretically, with a sufficient number of nodes in the hidden layer, any nonlinear function can be approximated
Correlated Variables • Variables used in model building are often correlated. • It is difficult to isolate the effect of the individual variables because of the correlation between the variables.
Correlated Variables: An Example • Workers Compensation Line • Produce an economic inflation index • Wage Inflation • Medical Inflation • Benefit Level Index • In simplified example no other variable drives severity results
Factor Analysis Example X1 = b1 Factor1 X2 = b2 Factor1 X3 = b3 Factor1 Index =.395 (Wage Inflation)+.498(Medical Inflation)+.113(Benefit Level Inflation)
Interpreting Neural Network • Look at weights to hidden layer • Compute sensitivities: • a measure of how much the predicted value’s error increases when the variables are excluded from the model one at a time
Table 9: Factor Example Parameters Table 10 W0 Sensitivities of Variables in Factor Example W1 W2 W3 Benefit Level 2.549 -2.802 23.6% -3.010 0.662 Medical Inflation 33.1% Wage Inflation 6.0% Interpretation of Neural Network
Interactions: Another Modeling Problem • Impact of two variables is more or less than the sum of their independent impacts.