# Difference between revisions of "General approach"

m |
m |
||

Line 3: | Line 3: | ||

[[File:Figure_3-1.JPG|thumb|600px|center|'''Figure 3.1.''' A generic algorithm for proceeding with an analytical project within the framework of Clinical Analytics.]] | [[File:Figure_3-1.JPG|thumb|600px|center|'''Figure 3.1.''' A generic algorithm for proceeding with an analytical project within the framework of Clinical Analytics.]] | ||

− | As mentioned in | + | As mentioned in [[Introduction]], the majority of predictive analytic problems can be solved by employing one of two wide types of forecasting methodologies: regression and classification. Regression<ref>Or another appropriate fitting technique |

</ref> should be used when the output variable<ref>Scalar, i.e., single-valued or vector, i.e., multivalued. | </ref> should be used when the output variable<ref>Scalar, i.e., single-valued or vector, i.e., multivalued. | ||

</ref> is interval or continuous, i.e., can take on any permissible value inside an interval (which may include the whole real axis). Examples of this type of problem include predicting: | </ref> is interval or continuous, i.e., can take on any permissible value inside an interval (which may include the whole real axis). Examples of this type of problem include predicting: |

## Latest revision as of 15:29, 27 June 2016

An outline of a general approach to solving an analytical problem is presented in Fig. 3.1.

As mentioned in Introduction, the majority of predictive analytic problems can be solved by employing one of two wide types of forecasting methodologies: regression and classification. Regression^{[1]} should be used when the output variable^{[2]} is interval or continuous, i.e., can take on any permissible value inside an interval (which may include the whole real axis). Examples of this type of problem include predicting:

- a lab test result based on the patient’s demographics, clinical history and other lab tests;
- the number of admissions based on the previous history and calendar data;
- patient management cost based on patient’s data.

Logistic regression is one of the most widely practically used classification algorithms. It is easy to implement^{[3]}, intuitive and can be made sufficiently accurate for most uncomplicated modeling tasks. Mathematically similar to linear regression, it^{[4]} can (and often should) be used when the output variable is an indicator, binary, categorical, nominal or ordinal. Examples of this type of problem include

- predicting patient’s risk of mortality, admission or readmission based on demographic and clinical data;
- classifying the severity of a patient’s condition based on available clinical data and history;
- identifying those patients among high risk population who are most likely to respond to intervention[3].