关于深度学习:ST300-Assessed

Week 15
ST300 Assessed Coursework
Body Measurements
1 Question
The data-set body.dat contains 21 body dimension measurements, together
with age, weight, height, and gender on 507 individuals. The 247 men and
260 women were primarily individuals in their twenties and thirties, with a
scattering of older men and women, all exercising several hours a week.
The main aim is to explain how the height of an individual is related to
other variables. Using R (other packages not allowed), obtain a satisfactory
model for height in terms of the other variables. You should examine outliers
and influential observations, and try transformations. You may also want to
try variable selections using techniques learnt and used in your lectures and
workshops. Why is your model good? From your model, what can you
say about the relationship between the height and other body measurements
or characteristics of a particular individual? Which body measurement(s)
are potentially more important in estimating a person’s height? Write a
summary of your analysis with no more than 6 sides of A4 (including
graphs and figures).
Variable descriptions:
Skeletal Measurements: (all in cm)
biacromial Biacromial diameter
pelvic.breadth Biiliac diameter, or “pelvic breadth”
bitrochanteric Bitrochanteric diameter, in cm
chest.depth Chest depth between spine and sternum at nipple level,
mid-expiration
chest.diam Chest diameter at nipple level, mid-expiration
elbow.diam Elbow diameter, sum of two elbows
wrist.diam Wrist diameter, sum of two wrists
knee.diam Knee diameter, sum of two knees
ankle.diam Ankle diameter, sum of two ankles
February 11, 2019 ST300 1
Week 15
Girth Measurements: (all in cm)
shoulder.girth Shoulder girth over deltoid muscles
chest.girth Chest girth, nipple line in males and just above breast
tissue in females, mid-expiration
waist.girth Waist girth, narrowest part of torso below the rib cage,
average of contracted and relaxed position
navel.girth Navel (or “Abdominal”) girth at umbilicus and iliac crest,
iliac crest as a landmark
hip.girth Hip girth at level of bitrochanteric diameter
thigh.girth Thigh girth below gluteal fold, average of right and left
girths
bicep.girth Bicep girth, flexed, average of right and left girths
forearm.girth Forearm girth, extended, palm up, average of right and
left girths
knee.girth Knee girth over patella, slightly flexed position, average
of right and left girths
calf.girth Calf maximum girth, average of right and left girths
ankle.girth Ankle minimum girth, average of right and left girths
wrist.girth Wrist minimum girth, average of right and left girths
Other Measurements:
age Age (years)
weight Weight (kg)
height Height (cm)
gender Gender (1 – male, 0 – female)
Note: Check all the covariates you are not familiar with before interpreting
your model. Note that some covariates are discrete, so that they can be
interpreted as factors. You are not taught about regression with a mixed
of factors and continuous covariates (called ANCOVA), but using the lm
function in R you can still get appropriate coefficients and anova function
with sum of squares calculated etc. But remember your ultimate goal is to
find a simple model for interpretation.
February 11, 2019 ST300 2
Week 15
2 Assessment
This question counts for 15% of the total mark for ST300. The work will be
subject to checks for plagiarism.
3 Submission
Your work should be submitted anonymously under your candidate
number. You can look up your candidate number in LSE for you.
A printed copy of the project must be put in the project drop-box
in the waiting area of the 6th floor of Columbia House by 12:00pm
on the 20th of March 2019. You must sign the plagiarism form
and put it separately in the department office collection point
of such form. In addition, you must upload your project using the
specified Moodle link on the ST300 Moodle page by 12:00pm on
the 20th of March 2019. The Moodle upload link will stop working
after that time and you will only be able to email me your project.
Penalties will apply for late submission of either the printed or the
electronic version of your project (but will not be doubly-penalised
if both are not submitted on time). For the plagiarism form see
the link in Moodle before the links to the project.
When the submission deadline has passed, a list of candidate numbers
for the work received will be put up on Moodle. Each student
must check from that list that their work has been received, counting
both the electronic and the printed versions.
Extensions to deadlines for coursework will only be given in fully
documented serious extenuating circumstances. There will be 5n
marks deducted from the total possible 100 marks for this coursework
if either the hard copy or the electronic version is submitted
between 24(n 1) and 24n hours from the stipulated deadline (including
working days only).
WX：codehelp