centercenterPredictive Modelling 216096843 Tutubala KZ8820090900Predictive Modelling 216096843 Tutubala KZ Table of Content 2 Predictive Modelling 2
centercenterPredictive Modelling
216096843
Tutubala KZ8820090900Predictive Modelling
216096843
Tutubala KZ
Table of Content
2 Predictive Modelling
2.1 Introduction Predictive Modelling
2.2 Advantages of Predictive Modelling
2.3 What does a typical predictive modelling process involve?
2.4 Predictive Modelling Process
2.5 What are the predictive modelling techniques?
2.6 Conclusion
2.1 Introduction Predictive Modelling
Predictive Modelling is the process of creating a model whose primary goal is to achieve high levels of accuracy. “Prediction models aim to accurately estimate the probability that a disease is present or that a future event will occur” (Shmuel, 2010; 25). In predictive modeling, overall predictive accuracy is paramount and the role of individual variables is less critical. Variables may be included in the final model even if they are not causally related to the outcome.
Predictive modelers need to consider several aspects of model performance. They also need to worry about overfitting and generalizability. Over fit models are tuned to the idiosyncrasies of a particular sample and thus have high predictive accuracy for the sample but not for new observations. Because of the problem of overfitting, prediction models should always undergo validation.
2.2 Advantages of Predictive Modelling
Deploying analytics for analyzing past, present and projected future outcome
Choosing the right step to drive the action in the most optimal manner
Predictive Analytics includes both decision optimization and advanced analytics
Supporting action and recommended actions are sent to the decision-makers
It helps to take proactive risk management measures
Testing iterative actions for the intended and unintended consequences
Process improvement, cost reduction and revenue generation are all possible
2.3 What does a typical predictive modelling process involve?
312674023495000Developing the right solutions to business problems using predictive modelling requires close collaboration with business subject matter experts from start to finish. The following are central to the process
Identify a problem where predictions of future outcomes or behavior can enhance the accuracy or efficiency of business decision-making.
Understand the business: know the products, the needs of the stakeholders, the resources and data available, identify assumptions and constraints, and the means of implementing a predictive modelling solution.
Clearly define the outcome to be predicted (the response variable) by the predictive model.
2.4 Predictive Modelling Process
2827020103441500Predictive modelling process involve running algorithm on data for prediction as the process is iterative it trains the model which gives the most fit knowledge for business fulfilment. Below are some of the stages of analytical modelling. Predictive Modelling is that data which is being generated in daily basis or the historical data may contain information for the present day business to get a maximum outcome with precision.
Data Gathering and Cleansing. Gather data from all the sources to extract needful information by cleansing operations to remove noisy data so that prediction can be accurate.
Data Analysis/Transformation. For normalization data need to be transformed for efficient processing. Scaling the values to a range normalization so that significance if data is not lost. Also remove irrelevant elements by correlation analysis to determine final outcome.
Building a Predictive Model: Predictive model uses regression technique to build predictive model by using classification algorithm. Identify test data and apply classification rules to check efficiency of classification model against test data.
Inferences/Evaluation: To make inferences perform cluster analysis and create data groups.
2.5 What are the predictive modelling techniques?
Predictive modelling techniques can be classified in a number of ways. The majority of the current applications in predictive modelling for insurance are based in supervised learning techniques. Within supervised learning, there are two primary subsets of model.
Classification models: In a classification problem, the objective is to predict a categorical outcome. This could include a binary response (i.e., 1 if policy lapsed; 0 if policy did not lapse), or could include a broad categorization/ multiple states.
Regression models: In a regression problem, the objective is to predict a continuous outcome; for example, the severity of an auto insurance collision claim. There are modelling techniques that can handle both classification and regression problems; however, some models are better suited for one or the other. For example, one can use a linear regression model effectively to predict a continuous variable (a regression problem), but it is not as effective when predicting a binary response variable
Unsupervised learning model: In an unsupervised learning problem, the modeler does not attempt to predict a certain outcome, but rather seeks to uncover latent structure or attributes within the data. For example, a modeler may analyze a company’s customer base to detect its major customer segments.
Within unsupervised learning there are two primary subsets of model:
Clustering models: In a clustering problem, the objective is to group the data into similar categories or clusters. Since this is unsupervised, clustering algorithms will attempt to find patterns in the underlying data that provide more information for the modeler.
Dimensionality reduction models: In a dimensionality reduction problem, the objective is to condense the number of variables that are being considered. Again, since this is unsupervised, these algorithms attempt to find the lowest number of variables that provide the highest amount of information.
Semi-supervised learning: In a semi-supervised learning problem, the modeler is likely faced with a data set where only a partial amount of the response variable is known. One option in this scenario is to use the unsupervised portion of the data to enhance a supervised model. For example, this might be a relevant technique if you are attempting to predict the effectiveness of a continuing sales strategy that has been in place for five years but only started gathering data on the sales results beginning last year
2.6 Creating a predictive model happens in five stages:
Thesis, data collection, data study, data validation, and updating of the model.
43624556007000
Reprinted with permission of on the risk, Journal of the Academy of Life Underwriting (www.ontherisk.com).
2.7Conclusion
Adopting predictive modelling capabilities into our practices is critical to our future offerings and key to our ability to thrive in a competitive marketplace. Reinsurers need to focus on how we can add value to our clients and answer the looming question, “How will this technology and information affect our price?”
Here are some final thoughts on predictive modelling:
Make predictive modelling a high-priority initiative
Continue current activity and plan for future predictive modelling work
Add to staff, skills and experience.
Reference:
J Barretina, G Caponigro, N Stransky, K Venkatesan (2014). Epoxy. online Eds.a.ebscohost.com.ezproxy.liv.ac.uk. Available at: http://eds.a.ebscohost.com.ezproxy.liv.ac.uk/eds/detail?sid=fa9392f5-3794-4dd7-a8bc-0f2d5915a373%40sessionmgr4005;vid=4;hid=4110;bdata=JnNpdGU9ZWRzLWxpdmUmc2NvcGU9c2l0ZQ%3d%3d#db=bth;AN=67324416 Accessed 21 Jun. 2014.
Clarke, R.T., Furse, M.T., Wright, J.F. and Moss, D. (1996). Derivation of a biological quality index for river sites: comparison of the observed with the expected fauna. Journal of Applied Statistics 23, 311-332.
Belbin, L. and McDonald, C. (1993). Comparing three classification strategies for use in ecology. Journal of Vegetation Science 4, 341-348.
Wright, J.F., Furse, M.T., Armitage, P.D., and Moss, D. (1993). New procedures for identifying running-water sites subject to environmental stress and for evaluating sites for conservation, based on the macroinvertebrate fauna. Archiv für Hydrobiologi 127, 319-326.
Institute of Freshwater Ecology (1991). Testing and further development of RIVPACS. Interim Report No. R;D Project 243. (National Rivers Authority: Bristol, U.K.)
http://www.dummies.com/how-to/content/how-to-choose-an-algorithm-for-apredictive-http://cdn.oreillystatic.com/en/assets/1/event/85/Best%20Practices%20for%20
Building%20and%20Deploying%20Predictive%20Models%20over%20Big0
Wright, J.F., Furse, M.T. and Armitage, P.D. (1994). Use of macroinvertebrate communities to detect environmental stress in running waters. In ‘Water quality and stress indicators in marine and freshwater ecosystems: linking levels of organisation – individuals, populations, communities.’ (Eds D.W. Sutcliffe.) pp. 15-34. (Freshwater Biological Association: Ambleside, U.K.)
Faith, D.P., Minchin, P.R. and Belbin, L. (1987). Compositional dissimilarity as a robust measure of ecological distance. Vegetatio 69, 57-68.
R. Schrijver, P. Berentsen, R. A. Groeneveld, A. Corporaal, and T. de Koeijer, “Development of nature-Predictive Modelling systems with an optimization model(2010:44)
