Data presentation and descriptive statistics -


Data presentation and descriptive statistics Paola Grosso SNE research group Today with Jeroen van der Ham as “special guest”

Instructions for use •  I do talk fast: –  Ask me to repeat if something is not clear; –  I made an effort to keep it ‘interesting’, but you are the ‘guinea pigs’…feedback is welcome!

•  You will not get a grade: –  But you will have to do some ‘work’;

•  3 for the price of 2 –  We will start slow and accelerate; –  We will (ambitiously?) cover lots of material; –  We will also use more than the standard two hours.

Sep.06 2010 - Slide 2


Why should you pay attention?

We are going to talk about “Data presentation, analysis and basic statistics”.

Your idea is? Sep.06 2010 - Slide 4

Our motivation We want to avoid to hear this from you.

1.  An essential component of scientific research; 2.  A must-have skill (!) of any master student and researcher (… but useful also in commercial/industry/business settings);

3.  It will help to communicate more effectively your results (incidentally, it also means higher grades during RPs).

Sep.06 2010 - Slide 5

How to conduct a scientific project   Research your topic   Make a hypothesis.   Write down your procedure. •  • 

         

Control sample Variables

Assemble your Materials. Conduct the experiment. Repeat the experiment. Analyze your results. Draw a Conclusion.

This is our main focus! Sep.06 2010 - Slide 6

Roadmap for today and next week •  •  •  • 

Collecting data Presenting data Descriptive statistics A real-life example (Jeroen)

•  •  •  •  • 

Basic probability theory Probability distributions Parameter estimation Confidence intervals, limits, significance Hypothesis testing

Sep.06 2010 - Slide 7

Collecting data

Terminology Sampling Data types

Basic terminology •  Population = the collection of items under investigation •  Sample = a representative subset of the population, used in the experiments

Estimate the height?

•  Variable = the attribute that varies in each experiment •  Observation = the value of a variable during taken during one of the experiments. Sep.06 2010 - Slide 9

Quick test

Estimate the proportion of a population given a sample.

The FNWI has N students: you interview n students on whether they use public transport to come to the Science Park; a students answer yes. Can you estimate the number of students who travel by public transport?

Sep.06 2010 - Slide 10

The problem of bias

Sep.06 2010 - Slide 11

Sampling •  Non-probability sampling: some elements of the population have no chance of selection, or where the probability of selection can't be accurately determined. –  Accidental (or convenience) Sampling; –  Quota Sampling; –  Purposive Sampling.

•  Probability sampling: every unit in the population has a chance (greater than zero) of being selected in the sample, and this probability can be accurately determined. –  Simple random sample –  Systematic random sample –  Stratified random sample –  Cluster sample Sep.06 2010 - Slide 12


The attribute that varies in each experiment.

Qualitative variables, cannot be assigned a numerical value. Quantitative variables, can be assigned a numerical value. •  Discrete data values are distinct and separate, i.e. they can be counted

•  Categorical data values can be sorted according to category.

•  Nominal data values can be assigned a code in the form of a number, where the numbers are simply labels

•  Ordinal data values can be ranked or have a rating scale attached

•  Continuous data Values may take on any value within a finite or infinite interval

Sep.06 2010 - Slide 13

Quick test Discrete or continuous? –  –  –  –  –  – 

The The The The The The

number of suitcases lost by an airline. height of apple trees. number of apples produced. number of green M&M's in a bag. time it takes for a hard disk to fail. production of cauliflower by weight.

Sep.06 2010 - Slide 14

Presenting the data

Tables Charts Graphs

How many friends do you have on Facebook? …. 23,44,156,246,37,79,156,123,267,12, 145,88,95,156,32,287,167,55,256,47,

Frequency tables

•  A way to summarize data. •  It records how often each value of the variable occurs. How you build it? –  –  –  – 

Identify lower and upper limits Number of classes and width Segment data in classes Each value should fit in one (and no more) than one class: classes are mutually exclusive



Relative Frequency

Percentage (%)

Cumulative (less than)

Cumulative (greater than)




































3 Sep.06 2010 - Slide 16

Of course not everybody is a believer: “As the Chinese say, 1001 words is worth more than a picture” John McCartey

Sep.06 2010 - Slide 17

Histograms •  The graphical representation of a frequency table; •  Summarizes categorical, nominal and ordinal data; •  Display bar vertically or horizontally, where the area is proportional to the frequency of the observations falling into that class.

Useful when dealing with large data sets; Show outliers and gaps in the data set;

Sep.06 2010 - Slide 18

Building an histogram

Add values

Add title (or caption in document)

Add axis legends

Sep.06 2010 - Slide 19

Pie charts Suitable to represent categorical data; Used to show percentages; Areas are proportional to value of category.

Caution: • You should never use a pie chart to show historical data over time; • Also do not use for the data in the frequency distribution.

Sep.06 2010 - Slide 20

Line charts Are commonly used to show changes in data over time; Can show trends or changes well. Year

RP2 thesis

















Sep.06 2010 - Slide 21

Dependent vs. independent variables •  N.b= the terms are used differently in statistics than in mathematics! •  In statistics, the dependent variable is the event studied and expected to change whenever the independent variable is altered. •  The ultimate goal of every research or scientific analysis is to find relations between variables. Sep.06 2010 - Slide 22

Scatter plots •  Displays values for two variables for a set of data; •  The independent variable is plotted on the horizontal axis, the dependent variable on the vertical axis; •  It allows to determine correlation –  Positive (bottom left -> top right) –  Negative (top left -> bottom right) –  Null

with a trend line ‘drawn’ on the data.

Sep.06 2010 - Slide 23

Forest plot

… and more Shmoo plot

Bode plot Stemplot Arrhenius plot

Ternary plot

Bland-Altman plot

Galbraith plot

Recurrence plot Nichols plot Nyquist plot Lineweaver–Burk plot

Star plot

Funnel plot Violin plot Q-Q plot

Sep.06 2010 - Slide 24

Statistics packages followed by some hands on work

Graphics and statistics tools Plenty of tools to use to plot and do statistical analysis. Just some you could use: •  gnuplot •  ROOT •  Excel We will use the open-source statistical computer program R. Make installation yourself; $> apt-get install r-base-core

Run R as: $> R You find the documentation at: Sep.06 2010 - Slide 26

Quick exercise Create a CSV file with frequency data. Now in R: > > > > > >

salaries <- read.csv(file=”Path-to-file/Salary.csv") salaries salaries$Salary barplot(salaries$Salary) dev.copy(png,’MyBarPlot.png’)

Student,Salary 1,1250 2,2200 3,2345 4,6700 5,15000 6,3300 7,2230 8,1750 9,1900 10,1750 11,2100 12,2050

Can you improve this barplot? help(barplot) ??plot

Sep.06 2010 - Slide 27

Descriptive statistics

•  •  •  •  • 

Median, mean and mode Variance and standard deviation Basic concepts of distribution Correlation Linear regression

Median, mean and mode To estimate the centre of a set of observations, to convey a ‘one-liner’ information about your measurements, you often talk of average. Let’s be precise. Given a set of measurements: { x1, x2, …, xN} •  The median is the middle number in the ordered data set; below and above the median there is an equal number of observations. •  The (arithmetic) mean is the sum of the observations divided by the number of observations. :

•  The mode is the most frequently occurring value in the data set.

Sep.06 2010 - Slide 29

Quick test Look at the (fictitious!) monthly salary distribution of fresh OS3 graduates: What is median, mean and mode of this data set? Can you figure out how to do this in R?

What did you learn?

OS3 graduates

Monthly salary (gross in €)

Grad 1


Grad 2


Grad 3


Grad 4


Grad 5


Grad 6


Grad 7


Grad 8


Grad 9


Grad 10


Grad 11


Grad 12


Sep.06 2010 - Slide 30

Outliers •  An outlying observation is an observation that is numerically distant from the rest of the data (for example unusually large or small compared to others)

Causes: •  measurement error •  the population has a heavy-tailed distribution

Sep.06 2010 - Slide 31

Symmetry and skewness • 


A symmetrical distribution has the same number of values above and below the mean which is represented by the peak of the curve. The mean and median in a symmetrical distribution are equal.

Outliers create skewed distributions: •  Positively skewed if the outliers are above the mean: the mean is greater than the median and the mode; •  Negatively skewed if the outliers are below the mean: the mean is smaller than the median and the mode.

Dispersion and variability The mean represents the ‘central tendency’ of the data set. But alone it does not really gives us an idea of how the data is distributed. We want to have indications of the data variability. •  The range is the difference between the highest and lowest values in a set of data. It is the crudest measure of dispersion. •  The variance V(x) of x expresses how much x is liable to vary from its mean value: V (x) = =

1 (x i − x) 2 ∑ N i x2 − x


•  The standard deviation is the square root of the variance:

s€ x = V (x) =

1 2 2 2 (x − x) = x − x ∑ i N i

Sep.06 2010 - Slide 33

Different definitions of the Standard Deviation sx =

1 2 (x − x ) ∑ N i

is the S.D. of the data sample

•  Presumably our data was taken from a parent distributions which has mean µ and S.F. σ

Data Sample


Parent Distribution (from which data sample was drawn)

σ x


x – mean of our sample

µ – mean of our parent dist

s – S.D. of our sample

σ – S.D. of our parent dist

Beware Notational Confusion!

Sep.06 2010 - Slide 34

Different definitions of the Standard Deviation •  Which definition of σ you use, sdata or σparent, is matter of preference, but be clear which one you mean!

Data Sample

Parent Distribution (from which data sample was drawn)

sdata x

σparent µ

•  In addition, you can get an unbiased estimate of σparent from a given data sample using

σˆ parent =

1 N 2 (x − x ) = sdata ∑ N −1 i N −1

  sdata = 

 1 2 (x − x )  ∑ N i 

Quartiles and percentiles Quartiles: Q1, Q2 and Q3 divide the sample of observations into four groups: 25% of data points ≤ Q1; 50% of data points ≤ Q2; (Q2 is the median); 75% of data points ≤ Q3.

The semi-inter-quartile range (SIQR) , or quartile deviation, is:


Q3 − Q1 2

The 5-number summary: (min_value, Q1, Q2 , Q3 and max_value)

Percentiles: The values that divide the data sample in 100 equal parts. Sep.06 2010 - Slide 36

Box and whisker plot It uses the 5-number summary.

Sep.06 2010 - Slide 37

Correlation and regression

Correlation Correlation offers a predictive relationship that can be exploited in practice; it determines the extent to which values of the two variables are "proportional" to each other. .

Proportional means linearly related; that is, the correlation is high if it can be "summarized" by a straight line (sloped upwards or downwards); This line is called the regression line or least squares line, because it is determined such that the sum of the squared distances of all the data points from the line is the lowest possible. Sep.06 2010 - Slide 39

Covariance and Pearson’s correlation factor •  Given 2 variables x,y and a a dataset consisting of pairs of numbers: { (x1,y1), (x2,y2), … (xN,yN) } Dependencies between x and y are described by the sample covariance:

(has dimension D(x)D(y))

The sample correlation coefficient is defined as:


cov(x, y) sx sy

∈ [−1,+1]

(is dimensionless) Sep.06 2010 - Slide 40

Visualization of correlation r=0

r = 0.1

r = 0.5

r = -0.7

r = -0.9

r = 0.99

Correlation & covariance in >2 variables •  Concept of covariance, correlation is easily extended to arbitrary number of variables

•  so that a n x n symmetric matrix

takes the form of

•  This is called the covariance matrix, or error matrix •  Similarly the correlation matrix becomes

Sep.06 2010 - Slide 42

Quick test Create a CSV file with frequency data. Read the file into the R memory in variable obesity. Run the following commands:

Weight, Food_consumption 84,32 93,33 81,33 61,24 95,39 86,32 90,34 78,28 85,33 72,27 65,26 75,29

attach(obesity) plot(Weight,Food_consumption) cor(Weight,Food_consumption) cor(obesity) cor.test(Weight,Food_consumption)

What have you learned?

Sep.06 2010 - Slide 43

Careful with correlation coefficients!

•  Correlation does not imply cause •  Correlation is a measure of linear relation only •  Misleading influence of a third variable •  Spurious correlation of a part with the whole •  Combination of unlike population •  Inference to an unlike population

Sep.06 2010 - Slide 44

Least-square regression The goal is to fit a line to (xi,yi):

y i = a + bx i + εi such that the vertical distances εi (the error on yi) are minimized.

εi = y i − yˆ i

The resulting equation and coefficients are:

yˆ = a + bx

(x − x )(y − y ) cov(x, y) s ∑ b= = =r s s ∑ (x − x ) i






a = y − bx



Note, the correlation coefficient here

Sep.06 2010 - Slide 45

Quick test From the example before in R: > > > > > >

pairs(obesity) fit <- lm(Food_Consumption~Weight) fit summary(fit) plot(Weight,Food_consumption,pch=16) abline(lm(Food_consumption~Weight),col='red')

Sep.06 2010 - Slide 46

Jeroen van der Ham: An end-to-end statistical analysis

See you next week…


Data presentation and descriptive statistics -

Data presentation and descriptive statistics Paola Grosso SNE research group Today with Jeroen van der Ham as “special guest” Instructions for use •...

2MB Sizes 49 Downloads 17 Views

Recommend Documents

Descriptive Statistics and Descriptive Statistics and Exploratory Data
Jan 14, 2008 - What is descriptive statistics? Descriptive statistics describes describes your data. Visual and Numerica

Descriptive Statistics and Interpreting Statistics - Statistics Solutions
Descriptive Statistics and Interpreting Statistics Resources. Bartz, A. E. (1971). Basic descriptive statistics for educ

Visualization and descriptive statistics
Interesting analogy. • Blackett's reasoning about submarine sightings in WWII. • can estimate probability of sightin

Descriptive Statistics (Statistics Toolbox)
Empirical Cumulative Distribution Function. The ksdensity function described in the last section produces an empirical v

Understanding Your Data (Descriptive Statistics - Flinders University
1. Pawel Skuza 2013. Intermediate IBM SPSS. Understanding Your Data. (Descriptive Statistics, Graphs and Custom Tables).

Descriptive Statistics
Uncertainty in Engineering. • Uncertainty is inherent in all real world problems. • Two types of uncertainty: – Na

Descriptive Statistics
Descriptive Statistics. 6-1 Numerical Summaries of 6-4 Box Plots. CHAPTER OUTLINE. Chapter 6 Title and Outline. 1. 6-1 N

Descriptive Statistics
Descriptive Statistics. Y520. Robert S Michael. Goal: Learn to calculate indicators and construct graphs that summarize

Descriptive Statistics
The Analyst Application provides several types of descriptive statis- tics and graphical displays. The Summary Statistic

Descriptive Statistics
distinguished from inferential statistics. With descriptive statistics you are simply describing what the data shows. Wi