Graphical Elements

• Plots of point clouds for all "dependent" variables (x-axis) on tuberculosis (y-axis)
  • all countries (colored crosses)
  • only cluster 1 (red small dots)
• Variables colored according to all other graphics of en:Analysis of Tuberculosis (except: first plot, aids, usually red)
• Linear and nonlinear regression lines (red, solid) according to hypothetical relation
• Linear regression lines for cluster 1 (dashed, thin) according to hypothetical relation

Interpretation

Check en:Analysis of Tuberculosis.

Program Usage

If you want to display the point clouds and regression lines of different parts of the dataset in one plot, the code has to be slightly adjusted as exemplary done for three of the twelve elements of the graphic (little red dots, dashed lines).

Program Code

Attention! For repeating the computation a transformed dataset is needed! If you have not yet computated and saved the transformation, run the program for transformation on the wikipage en:Analysis of Tuberculosis first!

library("xplore")
library("stats")

; ----- Reading Data --------------------------------------------------------------------------

choose = "Read data from:"

defaults = "C:\Dokumente und Einstellungen\All Users\Desktop\UN_data_ordered.csv"

v1 = readvalue(choose, defaults)

x = readcsvm(v1)

x1 = readcsvm("C:\Dokumente und Einstellungen\All Users\Desktop\Cluster1.csv")
x1 = x1.double

data = x.double
country = x.text

; ----- 1.1) Nonlinear Regression of Aids on Tuberculosis --------------------------------------

ta = (0.8*log(data[,4]))~((data[,6]))

ta = paf(ta, isInf(ta[,1])==0)

{beta,bse,bstan,bpval} = linreg(ta[,1], ta[,2])

rx = #(0:max(data[,4]))

yq = beta[1] + beta[2]*log(rx)

rdata = sort (rx~yq)

rdata = setmask (rdata, "line", "red")

; ----- 4) Linear Regression of Sanitation on Tuberculosis -------------------------------------

gr4 = grlinreg(data[,8|6])

gr4 = setmask(gr4, "line", "red")

linreg(data[,8],data[,6])

; ----- 5) Linear Regression of Sanitation on Tuberculosis -------------------------------------

gr5 = grlinreg(data[,9|6])

gr5 = setmask(gr5, "line", "red")

linreg(data[,9],data[,6])

; ----- 5.1) Linear Regression (Cluster1) of Sanitation on Tuberculosis ------------------------

grcl5 = grlinreg(x1[,9|6])

grcl5 = setmask(grcl5, "line", "red", "size", "thin", "style", "dashed")

lr5cl1 = linreg(x1[,9],x1[,6])
lr5cl1

; ----- 6) Linear Regression of Sanitation on Tuberculosis -------------------------------------

gr6 = grlinreg(data[,10|6])

gr6 = setmask(gr6, "line", "red")

linreg(data[,10],data[,6])

; ----- 7) Linear Regression of Sanitation on Tuberculosis -------------------------------------

gr7 = grlinreg(data[,11|6])

gr7 = setmask(gr7, "line", "red")

linreg(data[,11],data[,6])

; ----- 8) Linear Regression of Sanitation on Tuberculosis -------------------------------------

gr8 = grlinreg(data[,12|6])

gr8 = setmask(gr8, "line", "red")

linreg(data[,12],data[,6])

; ----- 9) Linear Regression of 1/CO2^0.3 on Tuberculosis --------------------------------------

z9 = (data[,13]^(-0.3))~data[,6]

gr9 = grlinreg(z9)

gr9 = setmask(gr9, "line", "red")

; ----- 9.1) Nonlinear Regression --------------------------------------------------------------

{beta9,bse9,bstan9,bpval9} = linreg((data[,13]^(-0.3)), data[,6])

rx9 = #(0:max(data[,13])*100)/100

yq9 = beta9[1] + beta9[2]*rx9^(-0.3)

rdata9 = sort (rx9~yq9)

rdata9 = setmask (rdata9, "line", "red")

; ----- 10) Linear Regression of 1/Int^0.3 on Tuberculosis -------------------------------------

z10 = (data[,14]^(-0.3))~data[,6]

gr10 = grlinreg(z10)

gr10 = setmask(gr10, "line", "red")

; ----- 10.1) Nonlinear Regression -------------------------------------------------------------

{beta10,bse10,bstan10,bpval10} = linreg((data[,14]^(-0.3)), data[,6])

rx10 = #(0:max(data[,14])*1000)/1000

yq10 = beta10[1] + beta10[2]*rx10^(-0.3)

rdata10 = sort (rx10~yq10)

rdata10 = setmask (rdata10, "line", "red")

; ----- 10.2) Linear Regression (Cluster1) of Internet on Tuberculosis -------------------------

grcl10 = grlinreg(x1[,14|6])

grcl10 = setmask(grcl10, "line", "blue", "size", "thin", "style", "dashed")

lr10cl1 = linreg(x1[,14],x1[,6])
lr10cl1

; ----- 11) Linear Regression of 1/PC^0.3 on Tuberculosis --------------------------------------

z11 = (data[,15]^(-0.3))~data[,6]

gr11 = grlinreg(z11)

gr11 = setmask(gr11, "line", "red")

; ----- 11.1) Nonlinear Regression -------------------------------------------------------------

{beta11,bse11,bstan11,bpval11} = linreg((data[,15]^(-0.3)), data[,6])

rx11 = #(0:max(data[,15])*1000)/1000

yq11 = beta11[1] + beta11[2]*rx11^(-0.3)

rdata11 = sort (rx11~yq11)

rdata11 = setmask (rdata11, "line", "red")

; ----- 12) Linear Regression of 1/Tel^0.3 on Tuberculosis -------------------------------------

z12 = (data[,16]^(-0.3))~data[,6]

gr12 = grlinreg(z12)

gr12 = setmask(gr12, "line", "red")

; ----- 12.1) Nonlinear Regression -------------------------------------------------------------

{beta12,bse12,bstan12,bpval12} = linreg((data[,16]^(-0.3)), data[,6])

rx12 = #(0:max(data[,16])*300)/300

yq12 = beta12[1] + beta12[2]*rx12^(-0.3)

rdata12 = sort (rx12~yq12)

rdata12 = setmask (rdata12, "line", "red")

; ----- Create Colours -------------------------------------------------------------------------

randomize(25)
bunt = uniform(10, 3)*255
createcolor(bunt)

; ----- Graphical Settings ---------------------------------------------------------------------

setsize(800, 600)
f = createdisplay (3, 4)

; ----- Plotting the Point Clouds --------------------------------------------------------------

i = 4
j = 6

t1 = data[,i|j]
t2 = data[,i+1|j]
t3 = data[,i+3|j]
t4 = data[,i+4|j]
t5 = data[,i+5|j]
t6 = data[,i+6|j]
t7 = data[,i+7|j]
t8 = data[,i+8|j]
t9 = data[,i+9|j]
t10 = data[,i+10|j]
t11 = data[,i+11|j]
t12 = data[,i+12|j]

; ----- Example: Plotting the Cluster1 Point Cloud ---------------------------------------------

cl5 = x1[,i+5|j]
cl7 = x1[,i+7|j]
cl10 = x1[,i+10|j]

; ----- Graphical Settings ---------------------------------------------------------------------

setmaskp(t1, 0, 11)
setmaskp(t2, 5, 11)
setmaskp(t3, bunt[1], 11)
setmaskp(t4, bunt[2], 11)
setmaskp(t5, bunt[3], 11)
setmaskp(t6, bunt[4], 11)
setmaskp(t7, bunt[5], 11)
setmaskp(t8, bunt[6], 11)
setmaskp(t9, bunt[7], 11)
setmaskp(t10, bunt[8], 11)
setmaskp(t11, bunt[9], 11)
setmaskp(t12, bunt[10], 11)

setmaskp(cl5, 4, 1)
setmaskp(cl7, 4, 1)
setmaskp(cl10, 4, 1)

show(f, 1, 1, rdata, t1)
show(f, 1, 2, t2)
show(f, 1, 3, t3)
show(f, 1, 4, gr4, t4)

show(f, 2, 1, gr5, grcl5, t5, cl5)
show(f, 2, 2, gr6, t6)
show(f, 2, 3, gr7, t7, cl7)
show(f, 2, 4, gr8, t8)

setxaxis(f, 2, 1, min(min(data[,9:12])'), max(max(data[,9:12])'), 0)
setxaxis(f, 2, 2, min(min(data[,9:12])'), max(max(data[,9:12])'), 0)
setxaxis(f, 2, 3, min(min(data[,9:12])'), max(max(data[,9:12])'), 0)
setxaxis(f, 2, 4, min(min(data[,9:12])'), max(max(data[,9:12])'), 0)

show(f, 3, 1, rdata9, t9)
show(f, 3, 2, rdata10, grcl10, t10, cl10)
show(f, 3, 3, rdata11, t11)
show(f, 3, 4, rdata12, t12)

setyaxis(f, 3, 2, -25, 500)