DATA589/03_PPP_Models_Assignment.Rmd at main · NoonanM/DATA589 · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
---
title: "Assignment 03 - Fitting and Validating Poisson Point Process Models"
# author: "FisrtName LastName"
# date: "`r format(Sys.time(), '%d %B, %Y %H:%M:%OS')`"
output:
  html_document:
    css: tutorial.css
    fig_caption: yes
    highlight: textmate
    theme: flatly
    toc: yes
    toc_float: yes
---

```{r setup, include=FALSE}
# DO NOT ALTER CODE IN THIS CHUNK
knitr::opts_chunk$set(echo = TRUE, message = FALSE, fig.width=8, fig.height=6, fig.align = 'center')
```


* * *

Complete the following exercises before the submission deadline. In addition to the points detailed below, 5 points are assigned to the quality of the annotation, as well as to the 'cleanliness' of the code and resulting pdf document.

## Exercise 1 -- 1 point

We will again be working with the BC Parks dataset, which contains information on the locations of Provincial Parks in British Columbia. The parks belong to 5 different regions. There is also information on elevation (in m) and percent forest cover contained within the dataset.

* Import the BC park locations dataset and convert the data to a `ppp` object (for today you can exclude information on regions).  -- 1 point(s)

Note: You will need to load the `maptools` package and make use of the `as.owin()` function.

```{r}

```

## Exercise 2 -- 4 points

* Estimate and plot $\rho$ for the locations of parks as a function of both elevation and forest cover (be sure that the x-axis for elevation does not go below 0).  -- 2 point(s)
* Check for collinearity between elevation and forest cover (you will need to consider NA values).  -- 1 point(s)
* Based on these initial analyses, write down the expected form of the model. Provide justification for this starting point.  -- 1 point(s)

Note: Estimating rho can be slow ($\sim$ 1-2 min). Be sure to leave enough time for the document to knit.

```{r}

```

## Exercise 3 -- 4 points

* Fit the model you have defined in exercise 2 and inspect the model output. -- 1 point(s)
* Fit a null, intercept only model.  -- 1 point(s)
* Use AIC and a likelihood ratio test to determine if the model you defined is a better fit than the intercept only model.  -- 1 point(s)
* Write down the equation for the selected model. -- 0.5 point(s)
* Use this equation to estimate the intensity of parks at 500m elevation and 50% forest cover.

```{r}

```


## Exercise 4 -- 4 points

* Visualise the fitted model. Note: log scale the estimated intensity when plotting, ignore the standard error. You can use the `n` argument to adjust the resolution  -- 1 point(s)
* Plot the effects of the individual coefficients. Note: use the median value(s) of the other coefficients.  -- 2 point(s)
* Visually, do you think the model predictions are a good match to the data?  -- 1 point(s)

```{r}

```


## Exercise 5 -- 1 point

* Test whether the observed data deviate significantly from the model predictions. -- 1 point(s)

```{r}

```


## Exercise 5 -- 2 points

* Calculate and plot the model residuals.  -- 1 point(s)
* Based on the residuals, do you think the model performing well?  -- 1 point(s)

```{r}

```


## Exercise 6 -- 3 points

* Calculate the partial residuals as a function of both elevation and forest cover.  -- 1 point(s)
* Do you think that the terms are accurately capturing trends in the data?  -- 1 point(s)
* Do you have enough information to further refine the model and improve it's accuracy?  -- 1 point(s)

```{r}

```


## Exercise 7 -- 1 points

* Based on these analyses, what have you learned about the spatial distribution of parks in BC?  -- 1 point(s)

```{r}

```