Part 3 (4 points) Step 1. Predict the number of violent crimes in…

QuestionAnswered step-by-stepPart 3 (4 points) Step 1. Predict the number of violent crimes in… Part 3 (4 points)Step 1. Predict the number of violent crimes in communities by building three regression models. The variables included in the dataset involve the community, such as the percent of the population considered urban, and the median family income, and involving law enforcement, such as per capita number of police officers, and percent of officers assigned to drug units. Strategies for building a good model:Perform data preprocessing if needed (e.g., missing data imputation, variable transformation, data replacement, categorical data consolidation, etc.).Do not transform the target variable.Use variable selection using stepwise, forward, or backward regression techniques (with appropriate assessment criterion) if needed.Compare the three models’ performance (using model comparison node as illustrated in Part A of this assignment).Identifying variables (such as community name) that are not predictive and should not be used in the model. Target variable: NumViolentCrimesPerPop:  the number of violent crimes (including 8 crimes considered ‘Index Crimes’ by the FBI such as Murders, Robbery, etc.) per 100,000 population in a community.     Other Variables:Variable Description Typecommunityname  Community name – not predictive – for information only  string state  US state (by 2 letter postal abbreviation) nominal population  population for community numeric householdsize  mean people per household   numeric racepctblack  percentage of population that is african american  numeric racePctWhite  percentage of population that is caucasian  numeric racePctAsian  percentage of population that is of asian heritage  numeric racePctHisp  percentage of population that is of hispanic heritage  numeric agePct12t21  percentage of population that is 12-21 in age  numeric agePct12t29  percentage of population that is 12-29 in age  numeric agePct16t24  percentage of population that is 16-24 in age  numeric agePct65up  percentage of population that is 65 and over in age  numeric numbUrban  number of people living in areas classified as urban  numericpctUrban  percentage of people living in areas classified as urban  numeric medIncome  median household income  numeric pctWWage  percentage of households with wage or salary income two years ago  numeric pctWFarmSelf  percentage of households with farm or self employment income two years ago  numeric pctWInvInc  percentage of households with investment / rent income two years ago  numeric pctWSocSec  percentage of households with social security income two years ago  numeric pctWPubAsst  percentage of households with public assistance income two years ago  numeric pctWRetire  percentage of households with retirement income two years ago  numeric medFamInc  median family income. differs from household income for non-family households  numericperCapInc  per capita income  numeric whitePerCap  per capita income for caucasians  numeric blackPerCap  per capita income for african americans  numeric indianPerCap  per capita income for native americans  numeric AsianPerCap  per capita income for people with asian heritage  numeric OtherPerCap  per capita income for people with ‘other’ heritage  numeric HispPerCap  per capita income for people with hispanic heritage  numeric NumUnderPov  number of people under the poverty level  numeric PctPopUnderPov  percentage of people under the poverty level  numeric PctLess9thGrade  percentage of people 25 and over with less than a 9th grade education  numeric PctNotHSGrad  percentage of people 25 and over that are not high school graduates  numeric PctBSorMore  percentage of people 25 and over with a bachelor’s degree or higher education  numeric PctUnemployed  percentage of people 16 and over, in the labor force, and unemployed  numeric PctEmploy  percentage of people 16 and over who are employed  numeric PctEmplManu  percentage of people 16 and over who are employed in manufacturing  numeric PctEmplProfServ  percentage of people 16 and over who are employed in professional services  numeric PctOccupMgmtProf  percentage of people 16 and over who are employed in management or professional occupations  numeric MalePctDivorce  percentage of males who are divorced  numeric MalePctNevMarr  percentage of males who have never married  numeric FemalePctDiv  percentage of females who are divorced  numeric TotalPctDiv  percentage of population who are divorced  numeric PersPerFam  mean number of people per family  numeric PctFam2Par  percentage of families with kids that are headed by two parents  numericPctKids2Par  percentage of kids in family housing with two parents  numeric PctYoungKids2Par  percent of kids 4 and under in two parent households  numeric PctTeen2Par  percent of kids age 12-17 in two parent households  numeric PctWorkMomYoungKids  percentage of moms of kids 6 and under in labor force  numeric PctWorkMom  percentage of moms of kids under 18 in labor force  numeric NumKidsBornNeverMar  number of kids born to never married  numeric PctKidsBornNeverMar  percentage of kids born to never married  numeric NumImmig  total number of people known to be foreign born  numeric PctImmigRecent  percentage of _immigrants_ who immigated within last 3 years  numeric PctImmigRec5  percentage of _immigrants_ who immigated within last 5 years  numeric PctImmigRec8  percentage of _immigrants_ who immigated within last 8 years  numeric PctImmigRec10  percentage of _immigrants_ who immigated within last 10 years  numeric PctRecentImmig  percent of _population_ who have immigrated within the last 3 years  numeric PctRecImmig5  percent of _population_ who have immigrated within the last 5 years  numeric PctRecImmig8  percent of _population_ who have immigrated within the last 8 years  numeric PctRecImmig10  percent of _population_ who have immigrated within the last 10 years  numeric PctSpeakEnglOnly  percent of people who speak only English  numeric PctNotSpeakEnglWell  percent of people who do not speak English well  numeric PctLargHouseFam  percent of family households that are large (6 or more) numeric PctLargHouseOccup  percent of all occupied households that are large (6 or more people) numeric PersPerOccupHous  mean persons per household  numeric PersPerOwnOccHous  mean persons per owner occupied household  numeric PersPerRentOccHous  mean persons per rental household  numeric PctPersOwnOccup  percent of people in owner occupied households  numeric PctPersDenseHous  percent of persons in dense housing (more than 1 person per room) numeric PctHousLess3BR  percent of housing units with less than 3 bedrooms  numeric MedNumBR  median number of bedrooms   numeric HousVacant  number of vacant households  numeric PctHousOccup  percent of housing occupied  numeric PctHousOwnOcc  percent of households owner occupied  numeric PctVacantBoarded  percent of vacant housing that is boarded up  numeric PctVacMore6Mos  percent of vacant housing that has been vacant more than 6 months  numeric MedYrHousBuilt  median year housing units built  numericPctWOFullPlumb  percent of housing without complete plumbing facilities  numeric OwnOccLowQuart  owner occupied housing – lower quartile value  numeric OwnOccMedVal  owner occupied housing – median value  numeric OwnOccHiQuart  owner occupied housing – upper quartile value  numeric OwnOccQrange  owner occupied housing – difference between upper quartile and lower quartile values  numeric RentLowQ  rental housing – lower quartile rent  numeric RentMedian  rental housing – median rent  numeric RentHighQ  rental housing – upper quartile rent  numeric RentQrange  rental housing – difference between upper quartile and lower quartile rent  numeric MedRent  median gross rent  numeric MedRentPctHousInc  median gross rent as a percentage of household income  numeric MedOwnCostPctInc  median owners cost as a percentage of household income – for owners with a mortgage  numeric MedOwnCostPctIncNoMtg  median owners cost as a percentage of household income – for owners without a mortgage  numeric NumInShelters  number of people in homeless shelters  numeric NumStreet  number of homeless people counted in the street  numeric PctForeignBorn  percent of people foreign born  numeric PctBornSameState  percent of people born in the same state as currently living  numeric PctSameHouse5bf  percent of people living in the same house as 5 years before  numeric PctSameCity5bf  percent of people living in the same city as 5 years before  numeric PctSameState5bf  percent of people living in the same state as 5 years before  numeric LemasSwornFT  number of sworn full time police officers  numericLemasSwFTPerPop  sworn full time police officers per 100K population  numeric LemasSwFTFieldOps  number of sworn full time police officers in field operations (on the street as opposed to administrative etc) numeric LemasSwFTFieldPerPop  sworn full time police officers in field operations (on the street as opposed to administrative etc per 100K population) numeric LemasTotalReq  total requests for police   numeric LemasTotReqPerPop  total requests for police per 100K population  numeric PolicReqPerOffic  total requests for police per police officer  numeric PolicPerPop  police officers per 100K population  numeric RacialMatchCommPol  a measure of the racial match between the community and the police force. High values indicate proportions in community and police force are similar  numeric PctPolicWhite  percent of police that are caucasian  numeric PctPolicBlack  percent of police that are african american  numeric PctPolicHisp  percent of police that are hispanic  numeric PctPolicAsian  percent of police that are asian  numeric PctPolicMinor  percent of police that are minority of any kind  numeric OfficAssgnDrugUnits  number of officers assigned to special drug units  numeric NumKindsDrugsSeiz  number of different kinds of drugs seized  numeric PolicAveOTWorked  police average overtime worked   numeric LandArea  land area in square miles   numeric PopDens  population density in persons per square mile  numeric PctUsePubTrans  percent of people using public transit for commuting  numeric Which MODEL THAT HAS THE BEST PERFORMANCE:A screenshot of the diagram that shows all nodes from start to finish.Describe IN DETAIL the pre-processing steps you have performed (missing data, variable transformation, data replacement, nonnumeric data consolidation, etc.) and explain why.Provide screenshots of all variables’ roles and levels.For each node in your diagram: provide the settings you have used. (Don’t list all the settings. List only the settings you changed. If you didn’t change any settings for the node, indicate you used the default settings for all properties for the node).Report the “average squared error” value (for the validation set) of the model.Report the fit statistics of the model (F-value, p-value, and Adjusted R-square).List the variables, their estimates, and p-value (you can copy and paste from the appropriate section in the result window).Which variables are significant (p< 0.05) in explaining the target variable?Engineering & TechnologyComputer ScienceMIS 7620Share Question