# Multiple linear regression

# Import data

```r id=509f01f5-a364-4863-ab0a-5dd7be5d70e9
house <- read.csv('https://github.com/nchelaru/practical_statistics/raw/master/kc_house_data.csv')

house$date <- as.Date(house$date, "%Y%m%dT000000")

sapply(house, class)

head(house)
```

[__out-0.csv][nextjournal#output#509f01f5-a364-4863-ab0a-5dd7be5d70e9#__out-0.csv]

# Preview data

```r id=eb3fb79e-bffa-4117-b977-2bf41dae724d
#devtools::install_github("XanderHorn/autoEDA")

library(autoEDA)

autoEDA(house, y = 'price', IDFeats = 'id', sampleRate = 1, 
        outcomeType = "automatic", 
        maxUniques = 15, maxLevels = 25, removeConstant = TRUE, 
        removeZeroSpread = TRUE, removeMajorityMissing = TRUE, 
        imputeMissing = TRUE, clipOutliers = TRUE, minLevelPercentage = 0.025,
        predictivePower = TRUE, outlierMethod = "tukey", lowPercentile = 0.01,
        upPercentile = 0.99, plotCategorical = "stackedBar", 
        plotContinuous = "histogram", bins = 20, rotateLabels = FALSE, 
        colorTheme = 1, theme = 2, color = "#26A69A", transparency = 1, 
        outputPath = NULL, filename = "ExploratoryPlots", verbose = TRUE, 
        returnPlotList = FALSE)
```

![__out-1][nextjournal#output#eb3fb79e-bffa-4117-b977-2bf41dae724d#__out-1]

[__out-0.csv][nextjournal#output#eb3fb79e-bffa-4117-b977-2bf41dae724d#__out-0.csv]

# Multiple regression

The `jtools` package is a very useful one for summarizing and visualizing regression results. For more information, see [here](https://cran.r-project.org/web/packages/jtools/vignettes/summ.html).

```r id=eb08d714-7ea6-47cf-b889-94452a76e796
library(jtools)

house_lm <- lm(price ~ ., 
               data=house, 
               na.action=na.omit)

summ(house_lm)
```

# Stepwise regression

```r id=3a38093b-6125-457f-b438-92d44c7a2e06
library(MASS)

step <- stepAIC(house_lm, direction="both")

summ(step)
```

# Weighted regression

```r id=1ae3389b-357f-4299-b5f0-81b12c121c7c
library(lubridate)

house$Year = year(house$date)

house$Weight = house$Year - 2005

house_wt <- lm(price ~ ., 
               data=house, 
               weight=Weight)

summ(house_wt)
```

# Compare models

```r id=872cf33a-a813-408e-97eb-c84930c75412
plot_coefs(house_lm, house_wt, step, 
           scale = TRUE, 
           model.names = c("Full", "Weighted", "Stepwise"))
```

![__out-0.svg][nextjournal#output#872cf33a-a813-408e-97eb-c84930c75412#__out-0.svg]

[nextjournal#output#509f01f5-a364-4863-ab0a-5dd7be5d70e9#__out-0.csv]:
<https://nextjournal.com/data/QmTZjyiKKim8M8PJyNPfpEo8wDhT8vQcE4WzFY7MXnQ3tR?content-type=text/csv&node-id=509f01f5-a364-4863-ab0a-5dd7be5d70e9&filename=__out-0.csv&node-kind=output>

[nextjournal#output#eb3fb79e-bffa-4117-b977-2bf41dae724d#__out-1]:
<https://nextjournal.com/data/QmbLnpsemLuzQZvR31nUpRR6qgqVAx8oBzcPpBjTTuvfop?content-type=image/svg%2Bxml&node-id=eb3fb79e-bffa-4117-b977-2bf41dae724d&filename=__out-1&node-kind=output>

[nextjournal#output#eb3fb79e-bffa-4117-b977-2bf41dae724d#__out-0.csv]:
<https://nextjournal.com/data/QmQb5BTJWq6Jk6pY4hAs6wzxyMEqCXWAEwnYD4YjypLrHG?content-type=text/csv&node-id=eb3fb79e-bffa-4117-b977-2bf41dae724d&filename=__out-0.csv&node-kind=output>

[nextjournal#output#872cf33a-a813-408e-97eb-c84930c75412#__out-0.svg]:
<https://nextjournal.com/data/QmZTf44uVcyKZNfB3SdJ2uYr3q7sit5Ep9YiaBBZRyLzEu?content-type=image/svg%2Bxml&node-id=872cf33a-a813-408e-97eb-c84930c75412&filename=__out-0.svg&node-kind=output>

<details id="com.nextjournal.article">
<summary>This notebook was exported from <a href="https://nextjournal.com/a/M64Ao8YTRvLxWrsMKLboh?change-id=Cd5gpfJhrskYsqFVpyWcEU">https://nextjournal.com/a/M64Ao8YTRvLxWrsMKLboh?change-id=Cd5gpfJhrskYsqFVpyWcEU</a></summary>

```edn nextjournal-metadata
{:article
 {:settings nil,
  :nodes
  {"1ae3389b-357f-4299-b5f0-81b12c121c7c"
   {:compute-ref #uuid "c7dbb373-4dfa-4228-b681-f076dfad4a01",
    :id "1ae3389b-357f-4299-b5f0-81b12c121c7c",
    :kind "code",
    :runtime [:runtime "fbb04fb9-e412-4a90-a464-b3950395ffd6"],
    :stdout-collapsed? false},
   "3a38093b-6125-457f-b438-92d44c7a2e06"
   {:compute-ref #uuid "f7afa05a-6945-4652-87eb-3adc7e075edb",
    :id "3a38093b-6125-457f-b438-92d44c7a2e06",
    :kind "code",
    :runtime [:runtime "fbb04fb9-e412-4a90-a464-b3950395ffd6"]},
   "509f01f5-a364-4863-ab0a-5dd7be5d70e9"
   {:compute-ref #uuid "e8902164-a05a-42d6-b5f3-44c358c656bd",
    :exec-duration 547,
    :id "509f01f5-a364-4863-ab0a-5dd7be5d70e9",
    :kind "code",
    :output-log-lines {:stdout 1},
    :refs (),
    :runtime [:runtime "fbb04fb9-e412-4a90-a464-b3950395ffd6"]},
   "872cf33a-a813-408e-97eb-c84930c75412"
   {:compute-ref #uuid "f58c1e2a-64a2-4c25-b0f4-2333f003d42b",
    :id "872cf33a-a813-408e-97eb-c84930c75412",
    :kind "code",
    :runtime [:runtime "fbb04fb9-e412-4a90-a464-b3950395ffd6"]},
   "eb08d714-7ea6-47cf-b889-94452a76e796"
   {:compute-ref #uuid "ab726806-9f9a-48e8-a99f-ada7a9e10e24",
    :id "eb08d714-7ea6-47cf-b889-94452a76e796",
    :kind "code",
    :runtime [:runtime "fbb04fb9-e412-4a90-a464-b3950395ffd6"]},
   "eb3fb79e-bffa-4117-b977-2bf41dae724d"
   {:compute-ref #uuid "a6cb082b-f7fb-44a4-a6bd-c8b45a5c263b",
    :id "eb3fb79e-bffa-4117-b977-2bf41dae724d",
    :kind "code",
    :runtime [:runtime "fbb04fb9-e412-4a90-a464-b3950395ffd6"],
    :stdout-collapsed? true},
   "fbb04fb9-e412-4a90-a464-b3950395ffd6"
   {:environment
    [:environment
     {:article/nextjournal.id
      #uuid "5b45e6f7-fe51-488a-b89b-1c8f74dfb387",
      :change/nextjournal.id
      #uuid "5de55a70-27b5-48ac-a6aa-6aa24a73c87e",
      :node/id "4e307b44-52c3-4f79-9f3d-23047859e2fa"}],
    :environment? true,
    :id "fbb04fb9-e412-4a90-a464-b3950395ffd6",
    :kind "runtime",
    :language "r",
    :name "multiple_linear_regression",
    :type :nextjournal,
    :docker/environment-image
    "docker.nextjournal.com/environment@sha256:dbfe670eeaec504b2a8c1187a33841379978a16299acc880a70f368a742bfb2c"}},
  :nextjournal/id #uuid "02cdfd8c-d368-4b40-b6dd-f95f8170b7f4",
  :article/change
  {:nextjournal/id #uuid "5e1e306e-a68a-49be-a05d-e118a1ca3361"}}}

```
</details>