310.INTERNATIONAL SPACE MISSIONS ANALYSIS USING PROC SQL | PROC MEANS | PROC UNIVARIATE | PROC SGPLOT | MACRO-BASED SUCCESS RATE COMPUTATION

INTERNATIONAL SPACE MISSIONS ANALYSIS USING PROC SQL | PROC MEANS | PROC UNIVARIATE | PROC SGPLOT | MACRO-BASED SUCCESS RATE COMPUTATION

options nocenter validvarname=any nodate nonumber;

1) Create Space Missions Dataset

data work.space_missions;

    length Mission_ID $8 Country $20 Agency $30 Mission_Type $20 Outcome $12;

    infile datalines dlm=',' dsd truncover;

    input Mission_ID :$8.

          Country    :$20.

          Agency     :$30.

          Launch_Year

          Mission_Type :$20.

          Outcome      :$12.;

    datalines;

M001,USA,NASA,1998,Orbiter,Success

M002,Russia,Roscosmos,2000,Orbiter,Partial

M003,India,ISRO,2008,Orbiter,Success

M004,China,CNSA,2013,Lander,Success

M005,ESA,ESA,2014,Orbiter,Failure

M006,Japan,JAXA,2010,Probe,Success

M007,USA,SpaceX,2018,Orbiter,Success

M008,India,ISRO,2019,Orbiter,Failure

M009,Russia,Roscosmos,2016,Probe,Partial

M010,China,CNSA,2020,Orbiter,Success

M011,France,CNES,2002,Probe,Success

M012,SouthKorea,KARI,2021,Orbiter,Partial

;

run;

proc print data=work.space_missions;

run;

OUTPUT:

Obs	Mission_ID	Country	Agency	Mission_Type	Outcome	Launch_Year
1	M001	USA	NASA	Orbiter	Success	1998
2	M002	Russia	Roscosmos	Orbiter	Partial	2000
3	M003	India	ISRO	Orbiter	Success	2008
4	M004	China	CNSA	Lander	Success	2013
5	M005	ESA	ESA	Orbiter	Failure	2014
6	M006	Japan	JAXA	Probe	Success	2010
7	M007	USA	SpaceX	Orbiter	Success	2018
8	M008	India	ISRO	Orbiter	Failure	2019
9	M009	Russia	Roscosmos	Probe	Partial	2016
10	M010	China	CNSA	Orbiter	Success	2020
11	M011	France	CNES	Probe	Success	2002
12	M012	SouthKorea	KARI	Orbiter	Partial	2021

2) Create a formatted outcome flag numeric: 1=Success, 0=NotSuccess 

data work.space_missions_flag;

    set work.space_missions;

    if upcase(strip(Outcome)) = 'SUCCESS' then Success_Flag = 1;

    else Success_Flag = 0;

run;

proc print data=work.space_missions_flag noobs label;

    title "International Space Missions - dataset (with Success_Flag)";

    var Mission_ID Country Agency Launch_Year Mission_Type Outcome Success_Flag;

run;

OUTPUT:

International Space Missions - dataset (with Success_Flag)

Mission_ID	Country	Agency	Launch_Year	Mission_Type	Outcome	Success_Flag
M001	USA	NASA	1998	Orbiter	Success	1
M002	Russia	Roscosmos	2000	Orbiter	Partial	0
M003	India	ISRO	2008	Orbiter	Success	1
M004	China	CNSA	2013	Lander	Success	1
M005	ESA	ESA	2014	Orbiter	Failure	0
M006	Japan	JAXA	2010	Probe	Success	1
M007	USA	SpaceX	2018	Orbiter	Success	1
M008	India	ISRO	2019	Orbiter	Failure	0
M009	Russia	Roscosmos	2016	Probe	Partial	0
M010	China	CNSA	2020	Orbiter	Success	1
M011	France	CNES	2002	Probe	Success	1
M012	SouthKorea	KARI	2021	Orbiter	Partial	0

3) List missions launched after 2010 

proc sql;

    title "Missions Launched After 2010";

    select Mission_ID, Agency, Country, Launch_Year, Mission_Type, Outcome

    from work.space_missions

    where Launch_Year > 2010

    order by Launch_Year;

quit;

OUTPUT:

Missions Launched After 2010

Mission_ID	Agency	Country	Launch_Year	Mission_Type	Outcome
M004	CNSA	China	2013	Lander	Success
M005	ESA	ESA	2014	Orbiter	Failure
M009	Roscosmos	Russia	2016	Probe	Partial
M007	SpaceX	USA	2018	Orbiter	Success
M008	ISRO	India	2019	Orbiter	Failure
M010	CNSA	China	2020	Orbiter	Success
M012	KARI	SouthKorea	2021	Orbiter	Partial

4) Count of missions by Outcome 

proc sql;

    title "Count of Missions by Outcome";

    select Outcome, count(*) as Count

    from work.space_missions

    group by Outcome;

quit;

OUTPUT:

Count of Missions by Outcome

Outcome	Count
Failure	2
Partial	3
Success	7

5) Numeric summary by Agency 

proc means data=work.space_missions n mean median min max std vardef=DF;

    class Agency;

    var Launch_Year;

    title "Numeric Summary (Launch_Year) by Agency - PROC MEANS";

run;

OUTPUT:

Numeric Summary (Launch_Year) by Agency - PROC MEANS

The MEANS Procedure

Analysis Variable : Launch_Year
Agency	N Obs	N	Mean	Median	Minimum	Maximum	Std Dev
CNES	1	1	2002.00	2002.00	2002.00	2002.00	.
CNSA	2	2	2016.50	2016.50	2013.00	2020.00	4.9497475
ESA	1	1	2014.00	2014.00	2014.00	2014.00	.
ISRO	2	2	2013.50	2013.50	2008.00	2019.00	7.7781746
JAXA	1	1	2010.00	2010.00	2010.00	2010.00	.
KARI	1	1	2021.00	2021.00	2021.00	2021.00	.
NASA	1	1	1998.00	1998.00	1998.00	1998.00	.
Roscosmos	2	2	2008.00	2008.00	2000.00	2016.00	11.3137085
SpaceX	1	1	2018.00	2018.00	2018.00	2018.00	.

6) Distribution diagnostics (skewness, kurtosis, extreme values) 

proc univariate data=work.space_missions cibasic;

    var Launch_Year;

    title "Launch Year Distribution - PROC UNIVARIATE";

    histogram Launch_Year / normal;

    inset mean median std / position=ne;

run;

OUTPUT:

Launch Year Distribution - PROC UNIVARIATE

The UNIVARIATE Procedure

Variable: Launch_Year

Moments
N	12	Sum Weights	12
Mean	2011.58333	Sum Observations	24139
Std Deviation	8.027887	Variance	64.4469697
Skewness	-0.5870186	Kurtosis	-1.0485987
Uncorrected SS	48558319	Corrected SS	708.916667
Coeff Variation	0.39908299	Std Error Mean	2.31745136

Basic Statistical Measures
Location		Variability
Mean	2011.583	Std Deviation	8.02789
Median	2013.500	Variance	64.44697
Mode	.	Range	23.00000
		Interquartile Range	13.50000

Basic Confidence Limits Assuming Normality
Parameter	Estimate	95% Confidence Limits
Mean	2012	2006	2017
Std Deviation	8.02789	5.68692	13.63037
Variance	64.44697	32.34102	185.78706

Tests for Location: Mu0=0
Test	Statistic		p Value
Student's t	t	868.0153	Pr > |t|	<.0001
Sign	M	6	Pr >= |M|	0.0005
Signed Rank	S	39	Pr >= |S|	0.0005

Quantiles (Definition 5)
Level	Quantile
100% Max	2021.0
99%	2021.0
95%	2021.0
90%	2020.0
75% Q3	2018.5
50% Median	2013.5
25% Q1	2005.0
10%	2000.0
5%	1998.0
1%	1998.0
0% Min	1998.0

Extreme Observations
Lowest		Highest
Value	Obs	Value	Obs
1998	1	2016	9
2000	2	2018	7
2002	11	2019	8
2008	3	2020	10
2010	6	2021	12

Launch Year Distribution - PROC UNIVARIATE

The UNIVARIATE Procedure

Launch Year Distribution - PROC UNIVARIATE

The UNIVARIATE Procedure

Fitted Normal Distribution for Launch_Year

Parameters for Normal Distribution
Parameter	Symbol	Estimate
Mean	Mu	2011.583
Std Dev	Sigma	8.027887

Goodness-of-Fit Tests for Normal Distribution
Test	Statistic		p Value
Kolmogorov-Smirnov	D	0.15337026	Pr > D	>0.150
Cramer-von Mises	W-Sq	0.06082002	Pr > W-Sq	>0.250
Anderson-Darling	A-Sq	0.40999932	Pr > A-Sq	>0.250

Quantiles for Normal Distribution
Percent	Quantile
	Observed	Estimated
1.0	1998.00	1992.91
5.0	1998.00	1998.38
10.0	2000.00	2001.30
25.0	2005.00	2006.17
50.0	2013.50	2011.58
75.0	2018.50	2017.00
90.0	2020.00	2021.87
95.0	2021.00	2024.79
99.0	2021.00	2030.26

7) Bar chart: count of missions by Agency colored by Outcome (stacked)

proc sgplot data=work.space_missions;

    title "Count of Missions by Agency and Outcome (stacked)";

    vbar Agency / group=Outcome datalabel;

    xaxis display=(nolabel);

    yaxis label="Number of Missions";

OUTPUT:

8) Bar chart: success rate by Mission_Type 

proc sql;

    create table work.type_success as

    select Mission_Type,

           sum(case when upcase(Outcome) = 'SUCCESS' then 1 else 0 end) as Successes,

           count(*) as Total_Missions,

           calculated Successes / calculated Total_Missions as Success_Rate format=percent8.2

    from work.space_missions

    group by Mission_Type;

quit;

proc print data=work.type_success;

run;

OUTPUT:

Obs	Mission_Type	Successes	Total_Missions	Success_Rate
1	Lander	1	1	100.0%
2	Orbiter	4	8	50.00%
3	Probe	2	3	66.67%

proc sgplot data=work.type_success;

    title "Success Rate by Mission Type";

    vbar Mission_Type / response=Success_Rate stat=sum datalabel;

    yaxis label='Success Rate' values=(0 to 1 by 0.25) grid;

run;

OUTPUT:

9) Macro to compute success rates by arbitrary grouping variable

%success_rate macro:

    Parameters:

      - data: dataset name (lib.dataset) to analyze

      - group: variable to group by (unquoted)

      - outcomevar: the Outcome variable name (default Outcome)

      - successvalue: the value of outcome considered a success (default SUCCESS)

      - outds: optional name for the output summary table (default work.success_summary)

%macro success_rate(data=work.space_missions, group=Agency, outcomevar=Outcome, successvalue=SUCCESS, outds=work.success_summary);

    %local ds lib mem;

    /* parse libname.member if provided */

    %if %index(&data,.) %then %do;

        %let lib = %scan(&data,1,.);

        %let mem = %scan(&data,2,.);

    %end;

    %else %do;

        %let lib = work;

        %let mem = &data;

    %end;

    /* Use PROC SQL to compute successes, totals, and percent */

    proc sql;

        create table &outds as

        select &group as Group_Value length=50,

               sum(case when upcase(strip(&outcomevar)) = upcase("&successvalue") then 1 else 0 end) as Successes,

               count(*) as Total,

               calculated Successes / calculated Total as Success_Rate format=percent8.2

        from &data

        group by &group

        order by Success_Rate desc;

    quit;

    title "Success Rate Summary by &group";

    proc print data=&outds noobs label;

        var Group_Value Successes Total Success_Rate;

        label Group_Value="&group"

              Successes="Successful Missions"

              Total="Total Missions"

              Success_Rate="Success Rate";

    run;

%mend success_rate;

%success_rate(data=work.space_missions, group=Agency);

OUTPUT:

Success Rate Summary by Agency

Agency	Successful Missions	Total Missions	Success Rate
SpaceX	1	1	100.0%
JAXA	1	1	100.0%
CNES	1	1	100.0%
NASA	1	1	100.0%
CNSA	2	2	100.0%
ISRO	1	2	50.00%
ESA	0	1	0.00%
KARI	0	1	0.00%
Roscosmos	0	2	0.00%

%success_rate(data=work.space_missions, group=Country);
OUTPUT:

Success Rate Summary by Country

Country	Successful Missions	Total Missions	Success Rate
France	1	1	100.0%
Japan	1	1	100.0%
China	2	2	100.0%
USA	2	2	100.0%
India	1	2	50.00%
SouthKorea	0	1	0.00%
Russia	0	2	0.00%
ESA	0	1	0.00%

%success_rate(data=work.space_missions, group=Mission_Type, successvalue=Success);

OUTPUT:

Success Rate Summary by Mission_Type

Mission_Type	Successful Missions	Total Missions	Success Rate
Lander	1	1	100.0%
Probe	2	3	66.67%
Orbiter	4	8	50.00%

10) Use the macro output to create a plot of success rates by Agency 

proc sgplot data=work.success_summary;

    title "Success Rate by Agency (from macro output)";

    vbar Group_Value / response=Success_Rate stat=sum datalabel;

    xaxis label='Agency' fitpolicy=rotate;

    yaxis label='Success Rate' values=(0 to 1 by 0.2);

run;

OUTPUT:

To Visit My Previous Electronic Data Analysis:Click Here

To Visit My Previous IPS Toppers Dataset:Click Here

To Visit My Previous Global Money Loan Trends Dataset:Click Here

To Visit My Previous Statewise Population Analysis:Click Here

Follow Us On : 

 

--- FOLLOW OUR BLOG FOR MORE INFORMATION.

--->PLEASE DO COMMENTS AND SHARE OUR BLOG.