11-model-selection.Rmd

---
title: "Model Selection"
author: "Kundan K. Rao"
date: "01/12/2021"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

# Model selection
## Internal measure
```{r}
library(clValid)
# wine data set:
# Computing clValid
clmethods <- c("hierarchical","kmeans","pam")
intern <- clValid(data.scaled, nClust = 2:6,
clMethods = clmethods, validation = "internal",method="complete")
# Summary
summary(intern)
```

## stability
```{r}
# Stability measures
clmethods <- c("hierarchical","kmeans","pam")
stab <- clValid(data.scaled, nClust = 2:6, clMethods = clmethods,
validation = "stability",method="complete")
# Display only optimal Scores
optimalScores(stab)
```

# computing p-value for hirarchical clustering
```{r}
library(pvclust)
set.seed(123)
res.pv <- pvclust(data.scaled, method.dist="euclidean",
method.hclust="average", nboot = 100)
# Default plot
plot(res.pv, hang = -1, cex = 0.5)
pvrect(res.pv)
```
# computing p-value for hirarchical clustering
```{r}
library(pvclust)
set.seed(123)
res.pv <- pvclust(data.scaled, method.dist="euclidean",
method.hclust="complete", nboot = 100)
# Default plot
plot(res.pv, hang = -1, cex = 0.5)
pvrect(res.pv)
```

# computing p-value for hirarchical clustering
```{r}
library(pvclust)
set.seed(123)
res.pv <- pvclust(data.scaled, method.dist="euclidean",
method.hclust="ward", nboot = 100)
# Default plot
plot(res.pv, hang = -1, cex = 0.5)
pvrect(res.pv)
```

# computing p-value for hirarchical clustering
```{r}
library(pvclust)
set.seed(123)
res.pv <- pvclust(data.scaled, method.dist="cor",
method.hclust="average", nboot = 100)
# Default plot
plot(res.pv, hang = -1, cex = 0.5)
pvrect(res.pv)
```


# computing p-value for hirarchical clustering
```{r}
library(pvclust)
set.seed(123)
res.pv <- pvclust(data.scaled, method.dist="cor",
method.hclust="complete", nboot = 100)
# Default plot
plot(res.pv, hang = -1, cex = 0.5)
pvrect(res.pv)
```


# computing p-value for hirarchical clustering
```{r}
library(pvclust)
set.seed(123)
res.pv <- pvclust(data.scaled, method.dist="cor",
method.hclust="average", nboot = 100)
# Default plot
plot(res.pv, hang = -1, cex = 0.5)
pvrect(res.pv)
```