All these indexes are based on concordant and discordant pairs.
@@ -260,9 +260,9 @@
variable also is higher on the other variable, and a pair of observations
is discordant if the subject who is higher on one variable is lower on
the other variable.
- More formally, a pair (i,j), i=1, 2, ..., n is concordant if
- (x(i)-x(j)) $\times$ (y(i)-y(j))>0.
- It is discordant if (x(i)-x(j) ) $\times$ (y(i)-y(j))<0
+ More formally, a pair $(i,j)$, $i=1, 2, ..., n$ is concordant if
+ $(x(i)-x(j)) \times (y(i)-y(j))>0$.
+ It is discordant if $(x(i)-x(j) ) \times (y(i)-y(j))<0$.
Let $C$ be the total number of concordant pairs (concordances) and $D$
the total number of discordant pairs (discordances) . If $C > D$ the
variables have a positive association, but if $C < D$ then the variables
diff --git a/toolbox/helpfiles/FSDA/datasets_mv.html b/toolbox/helpfiles/FSDA/datasets_mv.html
index f3a087995..835bad8a0 100644
--- a/toolbox/helpfiles/FSDA/datasets_mv.html
+++ b/toolbox/helpfiles/FSDA/datasets_mv.html
@@ -211,6 +211,50 @@
Data matrices
- enterpreneous and skilled self-employed among those of working age.
+
+ | fat |
+ Phisical measurements of 251 males. The variables
+ are
+
+ -
+ body_fat: Percent body fat using Brozek's equation, 457/Density - 414.2
+ -
+body_fat_siri: Percent body fat using Siri's equation, 495/Density - 450
+ -
+density: Density (gm/cm^2)
+ -
+ age: Age (yrs)
+ -
+weight: Weight (lbs)
+ -
+height: Height (inches)
+ -
+BMI: Adiposity index = Weight/Height^2 (kg/m^2)
+ - ffweight: Fat Free Weight = (1 - fraction of body fat) * Weight, using Brozek's formula (lbs)
+ -
+neck: Neck circumference (cm)
+ -
+chest: Chest circumference (cm)
+ -
+abdomen: Abdomen circumference (cm) "at the umbilicus and level with the iliac crest"
+ -
+hip: Hip circumference (cm)
+ - thigh: Thigh circumference (cm)
+ -
+knee: Knee circumference (cm)
+ -
+ankle: Ankle circumference (cm)
+ -
+bicep: Extended biceps circumference (cm)
+ - forearm: Forearm circumference (cm)
+ -
+wrist: Wrist circumference (cm) "distal to the styloid processes"
Note that
+ observation 182 in the original dataset has been removed because it
+ reported a percent body fat estimate equal to 0. The purpose is to predict
+ body_fat from the other measurements.
The source of the
+ data is attributed to Dr. A. Garth Fisher, Human Performance Research
+ Center, Brigham Young University, Provo, Utah 84602, |
+
| fondi |
The fondi data set, introduced by Zani (2000),
@@ -234,12 +278,30 @@ Data matrices
which allows different transformations for positive and negative responses is neede to analyze these data.
|
-
+
| head |
The Swiss Heads dataset was introduced by B. Flury
and H. Riedwyl (1988). It contains information on six variables
describing the dimensions of the heads of 200 twenty year old Swiss
soldiers. |
+
+
+ | hprice |
+ Sales prices of 546 houses in the city of Windsor,
+ Ontario, Canada, during July, August and September, 1987.
The variables
+ are
- lotsize: the lot size of a property in square feet
-
+ bedrooms: number of bedrooms
- bathrms: number of full bathrooms
-
+ stories number of stories excluding basement
- driveway does the
+ house has a driveway?
- recroom does the house has a recreational
+ room?
- fullbase: does the house has a full finished basement?
-
+ gashw: does the house uses gas for hot water heating?
- airco: does
+ the house has central air conditioning?
- garagepl: number of garage
+ places
- prefarea: is the house located in the preferred neighbourhood
+ of the city?
-price: sale price of a house.
The reference is
+ Verbeek, Marno (2004) A Guide to Modern Econometrics, John Wiley and Sons,
+ chapter 3. Journal of Applied Econometrics data archive :
+ http://qed.econ.queensu.ca/jae/
+ . |
| milk |
@@ -435,8 +497,8 @@ Contingency tables
| SportHealth |
- The dataset SportHealth contains a contingency table between "Physical Activity Frequency" and self assesment "Quality of Life Ratings".
- The number of people interviewed is 303. |
+ The SportHealth dataset contains a contingency table between "Physical Activity Frequency" and self assesment "Quality of Life Ratings".
+ The number of people interviewed is 303. |
diff --git a/toolbox/helpfiles/FSDA/images/corrOrdinal_01.png b/toolbox/helpfiles/FSDA/images/corrOrdinal_01.png
index b6f274eec..13f7db578 100644
Binary files a/toolbox/helpfiles/FSDA/images/corrOrdinal_01.png and b/toolbox/helpfiles/FSDA/images/corrOrdinal_01.png differ
diff --git a/toolbox/helpfiles/FSDA/images/corrOrdinal_02.png b/toolbox/helpfiles/FSDA/images/corrOrdinal_02.png
index 3b3e51e6a..8db84c48f 100644
Binary files a/toolbox/helpfiles/FSDA/images/corrOrdinal_02.png and b/toolbox/helpfiles/FSDA/images/corrOrdinal_02.png differ
diff --git a/toolbox/helpfiles/FSDA/images/corrOrdinal_03.png b/toolbox/helpfiles/FSDA/images/corrOrdinal_03.png
index 816708e4a..04025271e 100644
Binary files a/toolbox/helpfiles/FSDA/images/corrOrdinal_03.png and b/toolbox/helpfiles/FSDA/images/corrOrdinal_03.png differ
diff --git a/toolbox/helpfiles/FSDA/images/corrOrdinal_04.png b/toolbox/helpfiles/FSDA/images/corrOrdinal_04.png
index 4d6af91f4..825ff77d0 100644
Binary files a/toolbox/helpfiles/FSDA/images/corrOrdinal_04.png and b/toolbox/helpfiles/FSDA/images/corrOrdinal_04.png differ
diff --git a/toolbox/multivariate/corrOrdinal.m b/toolbox/multivariate/corrOrdinal.m
index a8be8e426..75144c6cb 100644
--- a/toolbox/multivariate/corrOrdinal.m
+++ b/toolbox/multivariate/corrOrdinal.m
@@ -150,11 +150,11 @@
% (third column), upper confidence limit (fourth column).
% Note that the standard errors in this table are computed not
% assuming the null hypothesis of independence.
-% out.IndContr2CminusD = IxJ array containing individual contributions to
-% the commpon numerator of all the indexes above (namely
+% out.Contrib2CminusD = IxJ array containing individual contributions to
+% the common numerator of all the indexes above (namely
% C-D).
-% out.IndContr2CminusDtable = IxJ table containing individual contributions to
-% the commpon numerator of all the indexes above (namely
+% out.Contrib2CminusDtable = IxJ table containing individual contributions to
+% the common numerator of all the indexes above (namely
% C-D).
%
%
@@ -166,9 +166,9 @@
% variable also is higher on the other variable, and a pair of observations
% is discordant if the subject who is higher on one variable is lower on
% the other variable.
-% More formally, a pair (i,j), i=1, 2, ..., n is concordant if
-% (x(i)-x(j)) $\times$ (y(i)-y(j))>0.
-% It is discordant if (x(i)-x(j) ) $\times$ (y(i)-y(j))<0
+% More formally, a pair $(i,j)$, $i=1, 2, ..., n$ is concordant if
+% $(x(i)-x(j)) \times (y(i)-y(j))>0$.
+% It is discordant if $(x(i)-x(j) ) \times (y(i)-y(j))<0$.
% Let $C$ be the total number of concordant pairs (concordances) and $D$
% the total number of discordant pairs (discordances) . If $C > D$ the
% variables have a positive association, but if $C < D$ then the variables
@@ -500,7 +500,7 @@
%{
%% Example 2 of use of option plots.
- % Opinion on the movied watched and age interval
+ % Opinion on the movie watched and age interval
load cinema.mat
out=corrOrdinal(cinema,'plots',true);
% It is clear the negative relationship between
@@ -746,7 +746,7 @@
pvaltaua = 2*(1 - normcdf(abs(ztaua))); %p-value (two-sided)
%% tau-b statistic
- % For computationl purposes it is better to use relative frequencies
+ % For computational purposes it is better to use relative frequencies
% rather than absolute frequencies
Pi=N/n; % matrix of relative frequencies
pdiff=(con-dis)/n;
@@ -755,7 +755,7 @@
delta2=sqrt(1 - sum((ndotj/n).^2));
tauij=(2 * pdiff + Pdiff * repmat(ndotj/n,I,1) ) * delta2 * delta1 + ...
(Pdiff * repmat(nidot/n,1,J) * delta2)/delta1;
- % setaub = standard errot used to compute the confidence interval
+ % setaub = standard error used to compute the confidence interval
setaub= sqrt(( ( sum(Pi(:) .* tauij(:).^2) - sum(Pi(:) .* tauij(:)).^2)/(delta1 * delta2)^4)/n);
% The formula written in the help section (which uses the absolute
@@ -788,7 +788,7 @@
pvaltauc = 2*(1 - normcdf(abs(ztauc))); %p-value (two-sided)
%% Somers' D statistic
- % Find standard error of Somers D stat
+ % Find standard error of Somers' D stat
nidotrep=repmat(nidot,1,J);
% sesom = standard errot used to compute the confidence interval
@@ -846,10 +846,10 @@
out.ConfLimtable=ConfLimtable;
NconMinusNdis=Ncon-Ndis;
-out.IndContr2CminusD=NconMinusNdis;
-IndContr2CminusDtable=Ntable;
-IndContr2CminusDtable{:,:}=NconMinusNdis;
-out.IndContr2CminusDtable=IndContr2CminusDtable;
+out.Contrib2CminusD=NconMinusNdis;
+Contrib2CminusDtable=Ntable;
+Contrib2CminusDtable{:,:}=NconMinusNdis;
+out.Contrib2CminusDtable=Contrib2CminusDtable;
if dispresults == true
if NoStandardErrors == false
Compare calculation of tau-b with that which comes from
Matlab function corr.
