* Encoding: UTF-8.

*Exercici 1

DATASET ACTIVATE DataSet1.
FREQUENCIES VARIABLES=death
  /ORDER=ANALYSIS.


*Seleccionar els morts.

USE ALL.
COMPUTE filter_$=(death="Yes").
VARIABLE LABELS filter_$ 'death="Yes" (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

*calculem mediana i rang anys jugats NBA

FREQUENCIES VARIABLES=anys_jugat_NBA
  /FORMAT=NOTABLE
  /STATISTICS=MINIMUM MAXIMUM MEDIAN
  /ORDER=ANALYSIS.

* Segmento per Posició.

SORT CASES  BY pos.
SPLIT FILE LAYERED BY pos.

FREQUENCIES VARIABLES=anys_jugat_NBA
  /FORMAT=NOTABLE
  /STATISTICS=MINIMUM MAXIMUM MEDIAN
  /ORDER=ANALYSIS.

* Desfiltrem i deseleccionem.

SPLIT FILE OFF.

FILTER OFF.
USE ALL.
EXECUTE.

*Calcular nova variable pts per game

COMPUTE pts_game=pts / g.
EXECUTE.

USE ALL.
COMPUTE filter_$=(etni2 = 0 & lefthanded="No").
VARIABLE LABELS filter_$ 'etni2 = 0 & lefthanded="No" (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

FREQUENCIES VARIABLES=pts_game
  /FORMAT=NOTABLE
  /STATISTICS=MAXIMUM MEDIAN
  /ORDER=ANALYSIS.

FILTER OFF.
USE ALL.
EXECUTE.

EXAMINE VARIABLES=cms BY etni2
  /PLOT BOXPLOT HISTOGRAM NPPLOT
  /COMPARE GROUPS
  /STATISTICS DESCRIPTIVES
  /CINTERVAL 95
  /MISSING LISTWISE
  /NOTOTAL.

* Calcul taules doble entrada o contingencia

CROSSTABS
  /TABLES=place2 BY death
  /FORMAT=AVALUE TABLES
  /CELLS=COUNT
  /COUNT ROUND CELL
  /BARCHART.

CROSSTABS
  /TABLES=place2 BY death
  /FORMAT=AVALUE TABLES
  /CELLS=COUNT ROW COLUMN TOTAL 
  /COUNT ROUND CELL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=pts_game MISSING=LISTWISE REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: pts_game=col(source(s), name("pts_game"))
  GUIDE: axis(dim(1), label("pts_game"))
  GUIDE: axis(dim(2), label("Frequency"))
  GUIDE: text.title(label("Simple Histogram of pts_game"))
  ELEMENT: interval(position(summary.count(bin.rect(pts_game))), shape.interior(shape.square))
END GPL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=pts_game etni2 MISSING=LISTWISE REPORTMISSING=NO 
    DATAFILTER=etni2(VALUES=ALL UNLABELED=INCLUDE) 
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: pts_game=col(source(s), name("pts_game"))
  GUIDE: axis(dim(1), label("pts_game"))
  GUIDE: axis(dim(2), label("Frequency"))
  GUIDE: text.title(label("Simple Histogram of pts_game"))
  GUIDE: text.footnote(label("Filtered by etni2 variable"))
  ELEMENT: interval(position(summary.count(bin.rect(pts_game))), shape.interior(shape.square))
END GPL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=pts_game death MISSING=LISTWISE REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: pts_game=col(source(s), name("pts_game"))
  DATA: death=col(source(s), name("death"), unit.category())
  GUIDE: axis(dim(1), label("pts_game"))
  GUIDE: axis(dim(2), label("Frequency"))
  GUIDE: legend(aesthetic(aesthetic.color.interior), label("death"))
  GUIDE: text.title(label("Stacked Histogram of pts_game by death"))
  ELEMENT: interval.stack(position(summary.count(bin.rect(pts_game))), color.interior(death), 
    shape.interior(shape.square))
END GPL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=death pts_game MISSING=LISTWISE REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: death=col(source(s), name("death"), unit.category())
  DATA: pts_game=col(source(s), name("pts_game"))
  DATA: id=col(source(s), name("$CASENUM"), unit.category())
  GUIDE: axis(dim(1), label("death"))
  GUIDE: axis(dim(2), label("pts_game"))
  GUIDE: text.title(label("Simple Boxplot of pts_game by death"))
  SCALE: linear(dim(2), include(0))
  ELEMENT: schema(position(bin.quantile.letter(death*pts_game)), label(id))
END GPL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=state COUNT()[name="COUNT"] place2 MISSING=LISTWISE 
    REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: state=col(source(s), name("state"), unit.category())
  DATA: COUNT=col(source(s), name("COUNT"))
  DATA: place2=col(source(s), name("place2"), unit.category())
  COORD: rect(dim(1,2), cluster(3,0))
  GUIDE: axis(dim(3), label("Former and active NBA players"))
  GUIDE: axis(dim(2), label("Count"))
  GUIDE: legend(aesthetic(aesthetic.color.interior), label("Place of birth"))
  GUIDE: text.title(label("Clustered Bar Count of Former and active NBA players by Place of birth"))    
  SCALE: linear(dim(2), include(0))
  ELEMENT: interval(position(place2*COUNT*state), color.interior(place2), 
    shape.interior(shape.square))
END GPL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=kilos cms MISSING=LISTWISE REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE
  /FITLINE TOTAL=NO SUBGROUP=NO.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: kilos=col(source(s), name("kilos"))
  DATA: cms=col(source(s), name("cms"))
  GUIDE: axis(dim(1), label("Weight—kg"))
  GUIDE: axis(dim(2), label("Height—cm"))
  GUIDE: text.title(label("Scatter Plot of Height—cm by Weight—kg"))
  ELEMENT: point(position(kilos*cms))
END GPL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=kilos cms death MISSING=LISTWISE REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE
  /FITLINE TOTAL=NO SUBGROUP=NO.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: kilos=col(source(s), name("kilos"))
  DATA: cms=col(source(s), name("cms"))
  DATA: death=col(source(s), name("death"), unit.category())
  GUIDE: axis(dim(1), label("Weight—kg"))
  GUIDE: axis(dim(2), label("Height—cm"))
  GUIDE: legend(aesthetic(aesthetic.color.interior), label("death"))
  GUIDE: text.title(label("Scatter Plot of Height—cm by Weight—kg by death"))
  ELEMENT: point(position(kilos*cms), color.interior(death))
END GPL.

* Chart Builder.
GGRAPH
  /GRAPHDATASET NAME="graphdataset" VARIABLES=kilos cms death anys_jugat_NBA MISSING=LISTWISE 
    REPORTMISSING=NO
  /GRAPHSPEC SOURCE=INLINE
   INLINETEMPLATE="<setStyle type='scatter'><style size='20pt'/></setStyle>"
  /FITLINE TOTAL=NO SUBGROUP=NO.
BEGIN GPL
  SOURCE: s=userSource(id("graphdataset"))
  DATA: kilos=col(source(s), name("kilos"))
  DATA: cms=col(source(s), name("cms"))
  DATA: death=col(source(s), name("death"), unit.category())
  DATA: anys_jugat_NBA=col(source(s), name("anys_jugat_NBA"))
  GUIDE: axis(dim(1), label("Weight—kg"))
  GUIDE: axis(dim(2), label("Height—cm"))
  GUIDE: legend(aesthetic(aesthetic.color.interior), label("death"))
  GUIDE: legend(aesthetic(aesthetic.size), label("anys_jugat_NBA"))
  GUIDE: text.title(label("Scatter Plot of Height—cm by Weight—kg by death by anys_jugat_NBA"))
  ELEMENT: point(position(kilos*cms), color.interior(death), size(anys_jugat_NBA))
END GPL.

DATASET DECLARE aggregate_unis.
SORT CASES BY college.
AGGREGATE
  /OUTFILE='aggregate_unis'
  /PRESORTED
  /BREAK=college
  /mp_median=MEDIAN(mp) 
  /mp_sum=SUM(mp)
  /N_BREAK=N.

* Export Output.
OUTPUT EXPORT
  /CONTENTS  EXPORT=ALL  LAYERS=PRINTSETTING  MODELVIEWS=PRINTSETTING
  /DOC  DOCUMENTFILE='F:\2017_18\Cursos\Curs_CIFE\2020_21\Practica1_Wrangling_Data\OUTPUT.doc'
     NOTESCAPTIONS=YES  WIDETABLES=WRAP PAGEBREAKS=YES
     PAGESIZE=MM(210.01999999999998, 297.01)  TOPMARGIN=MM(25.4)  
    BOTTOMMARGIN=MM(25.400000000000034)
     LEFTMARGIN=MM(25.4)  RIGHTMARGIN=MM(25.400000000000006).