How Fast Is Our Planet Warming? A Data-Driven Climate Change Study in SAS

options nocenter;

1) Create raw dataset with date formats

data climate_raw;

informat Observation_Date date9.;

format Observation_Date date9.;

input Country :$25. Observation_Date :date9. CO2_Emissions :8.2 /* metric tons per capita */

Average_Temperature :6.2 /*in °C */ Sea_Level_Rise :8.1 /* mm since baseline year */

Forest_Cover :6.2 /*in percent */ Pollution_Index :6.1;

datalines;

Sweden 01JAN2024 4.90 7.1 9.5 69.0 22.0

India 01JAN2024 1.90 26.8 18.2 24.0 65.5

USA 01JAN2024 14.50 12.3 22.4 33.9 40.2

Bangladesh 01JAN2024 0.50 25.1 35.6 17.0 78.9

Australia 01JAN2024 16.80 21.5 14.0 16.5 36.0

Brazil 01JAN2024 2.10 25.0 10.1 59.4 30.0

Germany 01JAN2024 8.40 10.5 8.8 31.9 28.5

China 01JAN2024 7.60 14.2 16.7 22.3 55.0

Maldives 01JAN2024 2.00 28.3 70.2 32.0 62.0

Ethiopia 01JAN2024 0.30 19.8 12.0 45.0 34.0

Japan 01JAN2024 9.20 13.6 19.1 67.0 29.8

UK 01JAN2024 5.90 10.8 11.4 12.8 35.3

;

run;

proc print data=climate_raw;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index
1	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0
2	01JAN2024	India	1.9	26.8	18.2	24.0	65.5
3	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2
4	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9
5	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0
6	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0
7	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5
8	01JAN2024	China	7.6	14.2	16.7	22.3	55.0
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0
10	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0
11	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8
12	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3

2) Add a baseline year and compute 'Years_Since_Baseline' using INTCK

data climate_raw2;

set climate_raw;

/* choose baseline date (e.g., 1990-01-01) for climate comparisons */

Baseline_Date = '01JAN1990'd;

format Baseline_Date date9.;

Years_Since_Baseline = intck('year', Baseline_Date, Observation_Date); /* integer years */

/* Also create a quarterly-aligned date using INTNX (e.g., shift to quarter start) */

Obs_QuarterStart = intnx('quarter', Observation_Date, 0, 'b'); /* beginning of quarter */

format Obs_QuarterStart date9.;

run;

proc print data=climate_raw2;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Baseline_Date	Years_Since_Baseline	Obs_QuarterStart
1	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0	01JAN1990	34	01JAN2024
2	01JAN2024	India	1.9	26.8	18.2	24.0	65.5	01JAN1990	34	01JAN2024
3	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2	01JAN1990	34	01JAN2024
4	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9	01JAN1990	34	01JAN2024
5	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0	01JAN1990	34	01JAN2024
6	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0	01JAN1990	34	01JAN2024
7	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5	01JAN1990	34	01JAN2024
8	01JAN2024	China	7.6	14.2	16.7	22.3	55.0	01JAN1990	34	01JAN2024
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0	01JAN1990	34	01JAN2024
10	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0	01JAN1990	34	01JAN2024
11	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8	01JAN1990	34	01JAN2024
12	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3	01JAN1990	34	01JAN2024

3) Create a small country_info dataset to demonstrate MERGE (e.g., Region)

data country_info;

length Country $25 Region $15 Income_Group $10;

infile datalines dlm='|';

input Country :$25. Region :$15. Income_Group :$10.;

datalines;

Sweden|Europe|High

India|Asia|Lower-Middle

USA|North America|High

Bangladesh|Asia|Low

Australia|Oceania|High

Brazil|South America|Upper-Middle

Germany|Europe|High

China|Asia|Upper-Middle

Maldives|Asia|Upper-Middle

Ethiopia|Africa|Low

Japan|Asia|High

UK|Europe|High

;

run;

proc print data=country_info;

run;

OUTPUT:


Obs	Country	Region	Income_Group
1	Sweden	Europe	High
2	India	Asia	Lower-Midd
3	USA	North America	High
4	Bangladesh	Asia	Low
5	Australia	Oceania	High
6	Brazil	South America	Upper-Midd
7	Germany	Europe	High
8	China	Asia	Upper-Midd
9	Maldives	Asia	Upper-Midd
10	Ethiopia	Africa	Low
11	Japan	Asia	High
12	UK	Europe	High

4) Sort datasets for MERGE by Country

proc sort data=climate_raw2;

by Country;

run;

proc print data=climate_raw2;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Baseline_Date	Years_Since_Baseline	Obs_QuarterStart
1	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0	01JAN1990	34	01JAN2024
2	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9	01JAN1990	34	01JAN2024
3	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0	01JAN1990	34	01JAN2024
4	01JAN2024	China	7.6	14.2	16.7	22.3	55.0	01JAN1990	34	01JAN2024
5	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0	01JAN1990	34	01JAN2024
6	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5	01JAN1990	34	01JAN2024
7	01JAN2024	India	1.9	26.8	18.2	24.0	65.5	01JAN1990	34	01JAN2024
8	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8	01JAN1990	34	01JAN2024
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0	01JAN1990	34	01JAN2024
10	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0	01JAN1990	34	01JAN2024
11	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3	01JAN1990	34	01JAN2024
12	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2	01JAN1990	34	01JAN2024

proc sort data=country_info;

by Country;

run;

proc print data=country_info;

run;

OUTPUT:


Obs	Country	Region	Income_Group
1	Australia	Oceania	High
2	Bangladesh	Asia	Low
3	Brazil	South America	Upper-Midd
4	China	Asia	Upper-Midd
5	Ethiopia	Africa	Low
6	Germany	Europe	High
7	India	Asia	Lower-Midd
8	Japan	Asia	High
9	Maldives	Asia	Upper-Midd
10	Sweden	Europe	High
11	UK	Europe	High
12	USA	North America	High

5) Merge climate indicators with country_info

data climate_merged;

merge climate_raw2 (in=a) country_info (in=b);

by Country;

if a; /* keep only climate records */

run;

proc print data=climate_merged;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Baseline_Date	Years_Since_Baseline	Obs_QuarterStart	Region	Income_Group
1	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0	01JAN1990	34	01JAN2024	Oceania	High
2	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9	01JAN1990	34	01JAN2024	Asia	Low
3	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0	01JAN1990	34	01JAN2024	South America	Upper-Midd
4	01JAN2024	China	7.6	14.2	16.7	22.3	55.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd
5	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0	01JAN1990	34	01JAN2024	Africa	Low
6	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5	01JAN1990	34	01JAN2024	Europe	High
7	01JAN2024	India	1.9	26.8	18.2	24.0	65.5	01JAN1990	34	01JAN2024	Asia	Lower-Midd
8	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8	01JAN1990	34	01JAN2024	Asia	High
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd
10	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0	01JAN1990	34	01JAN2024	Europe	High
11	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3	01JAN1990	34	01JAN2024	Europe	High
12	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2	01JAN1990	34	01JAN2024	North America	High

6) Compute a composite 'Risk_Score' using z-scores (standardization)

6a) Get means and std devs into a dataset via PROC MEANS

proc means data=climate_merged noprint;

var CO2_Emissions Average_Temperature Sea_Level_Rise Forest_Cover Pollution_Index;

output out=stats mean=mean_CO2 mean_Temp mean_SL mean_Forest mean_Poll

std=std_CO2 std_Temp std_SL std_Forest std_Poll;

run;

proc print data=stats;

run;

OUTPUT:


Obs	_TYPE_	_FREQ_	mean_CO2	mean_Temp	mean_SL	mean_Forest	mean_Poll	std_CO2	std_Temp	std_SL	std_Forest	std_Poll
1	0	12	6.175	17.9167	20.6667	35.9	43.1	5.38569	7.33322	17.2849	19.8330	17.8161

6b) Load stats into macro variables for later use

data _null_;

set stats;

call symputx('mean_CO2', mean_CO2);

call symputx('std_CO2', std_CO2);

call symputx('mean_Temp', mean_Temp);

call symputx('std_Temp', std_Temp);

call symputx('mean_SL', mean_SL);

call symputx('std_SL', std_SL);

call symputx('mean_Forest', mean_Forest);

call symputx('std_Forest', std_Forest);

call symputx('mean_Poll', mean_Poll);

call symputx('std_Poll', std_Poll);

run;

LOG:

 NOTE: There were 1 observations read from the data set WORK.STATS.

6c) Create z-scores and composite risk (weights can be customized)

data climate_scores;

set climate_merged;

/* avoid division by zero */

z_CO2 = (CO2_Emissions - &mean_CO2.) / ( &std_CO2. + 1e-8 );

z_Temp = (Average_Temperature - &mean_Temp.) / ( &std_Temp. + 1e-8 );

z_SL = (Sea_Level_Rise - &mean_SL.) / ( &std_SL. + 1e-8 );

/* lower Forest_Cover => higher risk, so invert */

z_Forest = -1 * (Forest_Cover - &mean_Forest.) / ( &std_Forest. + 1e-8 );

z_Poll = (Pollution_Index - &mean_Poll.) / ( &std_Poll. + 1e-8 );

/* Weighted composite score (weights sum to 1) */

Risk_Score = 0.30*z_CO2 + 0.25*z_Temp + 0.20*z_SL + 0.15*z_Forest + 0.10*z_Poll;

/* scale Risk_Score to 0-100 for readability */

minRisk = .; maxRisk = .;

run;

proc print data=climate_scores;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Baseline_Date	Years_Since_Baseline	Obs_QuarterStart	Region	Income_Group	z_CO2	z_Temp	z_SL	z_Forest	z_Poll	Risk_Score	minRisk	maxRisk
1	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0	01JAN1990	34	01JAN2024	Oceania	High	1.97282	0.48864	-0.38569	0.97817	-0.39852	0.74374	.	.
2	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9	01JAN1990	34	01JAN2024	Asia	Low	-1.05372	0.97956	0.86396	0.95296	2.00941	0.44545	.	.
3	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0	01JAN1990	34	01JAN2024	South America	Upper-Midd	-0.75663	0.96592	-0.61133	-1.18489	-0.73529	-0.35904	.	.
4	01JAN2024	China	7.6	14.2	16.7	22.3	55.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	0.26459	-0.50683	-0.22949	0.68572	0.66793	0.07642	.	.
5	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0	01JAN1990	34	01JAN2024	Africa	Low	-1.09085	0.25682	-0.50140	-0.45883	-0.51077	-0.48323	.	.
6	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5	01JAN1990	34	01JAN2024	Europe	High	0.41313	-1.01138	-0.68654	0.20168	-0.81948	-0.31791	.	.
7	01JAN2024	India	1.9	26.8	18.2	24.0	65.5	01JAN1990	34	01JAN2024	Asia	Lower-Midd	-0.79377	1.21138	-0.14271	0.60001	1.25729	0.25190	.	.
8	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8	01JAN1990	34	01JAN2024	Asia	High	0.56167	-0.58865	-0.09064	-1.56809	-0.74651	-0.30665	.	.
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	-0.77520	1.41593	2.86571	0.19664	1.06084	0.83014	.	.
10	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0	01JAN1990	34	01JAN2024	Europe	High	-0.23674	-1.47502	-0.64604	-1.66893	-1.18432	-0.93776	.	.
11	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3	01JAN1990	34	01JAN2024	Europe	High	-0.05106	-0.97047	-0.53611	1.16472	-0.43781	-0.23423	.	.
12	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2	01JAN1990	34	01JAN2024	North America	High	1.54576	-0.76592	0.10028	0.10084	-0.16277	0.29115	.	.

7) Determine percentile ranks using PROC RANK and then categorize risk using a macro

proc rank data=climate_scores out=climate_rank ties=mean percent;

var Risk_Score;

ranks Risk_Pct; /* 0-100 scale percentiles (0..99) */

run;

proc print data=climate_rank;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Baseline_Date	Years_Since_Baseline	Obs_QuarterStart	Region	Income_Group	z_CO2	z_Temp	z_SL	z_Forest	z_Poll	Risk_Score	minRisk	maxRisk	Risk_Pct
1	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0	01JAN1990	34	01JAN2024	Oceania	High	1.97282	0.48864	-0.38569	0.97817	-0.39852	0.74374	.	.	91.667
2	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9	01JAN1990	34	01JAN2024	Asia	Low	-1.05372	0.97956	0.86396	0.95296	2.00941	0.44545	.	.	83.333
3	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0	01JAN1990	34	01JAN2024	South America	Upper-Midd	-0.75663	0.96592	-0.61133	-1.18489	-0.73529	-0.35904	.	.	25.000
4	01JAN2024	China	7.6	14.2	16.7	22.3	55.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	0.26459	-0.50683	-0.22949	0.68572	0.66793	0.07642	.	.	58.333
5	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0	01JAN1990	34	01JAN2024	Africa	Low	-1.09085	0.25682	-0.50140	-0.45883	-0.51077	-0.48323	.	.	16.667
6	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5	01JAN1990	34	01JAN2024	Europe	High	0.41313	-1.01138	-0.68654	0.20168	-0.81948	-0.31791	.	.	33.333
7	01JAN2024	India	1.9	26.8	18.2	24.0	65.5	01JAN1990	34	01JAN2024	Asia	Lower-Midd	-0.79377	1.21138	-0.14271	0.60001	1.25729	0.25190	.	.	66.667
8	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8	01JAN1990	34	01JAN2024	Asia	High	0.56167	-0.58865	-0.09064	-1.56809	-0.74651	-0.30665	.	.	41.667
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	-0.77520	1.41593	2.86571	0.19664	1.06084	0.83014	.	.	100.000
10	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0	01JAN1990	34	01JAN2024	Europe	High	-0.23674	-1.47502	-0.64604	-1.66893	-1.18432	-0.93776	.	.	8.333
11	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3	01JAN1990	34	01JAN2024	Europe	High	-0.05106	-0.97047	-0.53611	1.16472	-0.43781	-0.23423	.	.	50.000
12	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2	01JAN1990	34	01JAN2024	North America	High	1.54576	-0.76592	0.10028	0.10084	-0.16277	0.29115	.	.	75.000

7a) Macro to categorize into Low/Moderate/High/Severe using Risk_Pct

%macro categorize(in=, out=, pctvar=Risk_Pct);

data &out.;

set &in.;

length Risk_Category $10.;

if &pctvar. < 25 then Risk_Category = 'Low';

else if 25 <= &pctvar. < 50 then Risk_Category = 'Moderate';

else if 50 <= &pctvar. < 75 then Risk_Category = 'High';

else Risk_Category = 'Severe';

run;

proc print data=&out.;

run;

%mend categorize;

%categorize(in=climate_rank, out=climate_final, pctvar=Risk_Pct);

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Baseline_Date	Years_Since_Baseline	Obs_QuarterStart	Region	Income_Group	z_CO2	z_Temp	z_SL	z_Forest	z_Poll	Risk_Score	minRisk	maxRisk	Risk_Pct	Risk_Category
1	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0	01JAN1990	34	01JAN2024	Oceania	High	1.97282	0.48864	-0.38569	0.97817	-0.39852	0.74374	.	.	91.667	Severe
2	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9	01JAN1990	34	01JAN2024	Asia	Low	-1.05372	0.97956	0.86396	0.95296	2.00941	0.44545	.	.	83.333	Severe
3	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0	01JAN1990	34	01JAN2024	South America	Upper-Midd	-0.75663	0.96592	-0.61133	-1.18489	-0.73529	-0.35904	.	.	25.000	Moderate
4	01JAN2024	China	7.6	14.2	16.7	22.3	55.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	0.26459	-0.50683	-0.22949	0.68572	0.66793	0.07642	.	.	58.333	High
5	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0	01JAN1990	34	01JAN2024	Africa	Low	-1.09085	0.25682	-0.50140	-0.45883	-0.51077	-0.48323	.	.	16.667	Low
6	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5	01JAN1990	34	01JAN2024	Europe	High	0.41313	-1.01138	-0.68654	0.20168	-0.81948	-0.31791	.	.	33.333	Moderate
7	01JAN2024	India	1.9	26.8	18.2	24.0	65.5	01JAN1990	34	01JAN2024	Asia	Lower-Midd	-0.79377	1.21138	-0.14271	0.60001	1.25729	0.25190	.	.	66.667	High
8	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8	01JAN1990	34	01JAN2024	Asia	High	0.56167	-0.58865	-0.09064	-1.56809	-0.74651	-0.30665	.	.	41.667	Moderate
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	-0.77520	1.41593	2.86571	0.19664	1.06084	0.83014	.	.	100.000	Severe
10	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0	01JAN1990	34	01JAN2024	Europe	High	-0.23674	-1.47502	-0.64604	-1.66893	-1.18432	-0.93776	.	.	8.333	Low
11	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3	01JAN1990	34	01JAN2024	Europe	High	-0.05106	-0.97047	-0.53611	1.16472	-0.43781	-0.23423	.	.	50.000	High
12	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2	01JAN1990	34	01JAN2024	North America	High	1.54576	-0.76592	0.10028	0.10084	-0.16277	0.29115	.	.	75.000	Severe

8) Append an extra observation (simulating newly collected data) using PROC APPEND

data new_obs;

informat Observation_Date date9.;

format Observation_Date date9.;

Country = 'Fiji';

Observation_Date = '01JAN2024'd;

CO2_Emissions=1.70;

Average_Temperature=26.6;

Sea_Level_Rise=40.2;

Forest_Cover=50.0;

Pollution_Index=45.0;

run;

proc print data=new_obs;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index
1	01JAN2024	Fiji	1.7	26.6	40.2	50	45

/* Append to climate_final */

proc append base=climate_final

data=new_obs force;

run;

proc print data=climate_final ;

run;

OUTPUT:


Obs	Observation_Date	Country	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Baseline_Date	Years_Since_Baseline	Obs_QuarterStart	Region	Income_Group	z_CO2	z_Temp	z_SL	z_Forest	z_Poll	Risk_Score	minRisk	maxRisk	Risk_Pct	Risk_Category
1	01JAN2024	Australia	16.8	21.5	14.0	16.5	36.0	01JAN1990	34	01JAN2024	Oceania	High	1.97282	0.48864	-0.38569	0.97817	-0.39852	0.74374	.	.	91.667	Severe
2	01JAN2024	Bangladesh	0.5	25.1	35.6	17.0	78.9	01JAN1990	34	01JAN2024	Asia	Low	-1.05372	0.97956	0.86396	0.95296	2.00941	0.44545	.	.	83.333	Severe
3	01JAN2024	Brazil	2.1	25.0	10.1	59.4	30.0	01JAN1990	34	01JAN2024	South America	Upper-Midd	-0.75663	0.96592	-0.61133	-1.18489	-0.73529	-0.35904	.	.	25.000	Moderate
4	01JAN2024	China	7.6	14.2	16.7	22.3	55.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	0.26459	-0.50683	-0.22949	0.68572	0.66793	0.07642	.	.	58.333	High
5	01JAN2024	Ethiopia	0.3	19.8	12.0	45.0	34.0	01JAN1990	34	01JAN2024	Africa	Low	-1.09085	0.25682	-0.50140	-0.45883	-0.51077	-0.48323	.	.	16.667	Low
6	01JAN2024	Germany	8.4	10.5	8.8	31.9	28.5	01JAN1990	34	01JAN2024	Europe	High	0.41313	-1.01138	-0.68654	0.20168	-0.81948	-0.31791	.	.	33.333	Moderate
7	01JAN2024	India	1.9	26.8	18.2	24.0	65.5	01JAN1990	34	01JAN2024	Asia	Lower-Midd	-0.79377	1.21138	-0.14271	0.60001	1.25729	0.25190	.	.	66.667	High
8	01JAN2024	Japan	9.2	13.6	19.1	67.0	29.8	01JAN1990	34	01JAN2024	Asia	High	0.56167	-0.58865	-0.09064	-1.56809	-0.74651	-0.30665	.	.	41.667	Moderate
9	01JAN2024	Maldives	2.0	28.3	70.2	32.0	62.0	01JAN1990	34	01JAN2024	Asia	Upper-Midd	-0.77520	1.41593	2.86571	0.19664	1.06084	0.83014	.	.	100.000	Severe
10	01JAN2024	Sweden	4.9	7.1	9.5	69.0	22.0	01JAN1990	34	01JAN2024	Europe	High	-0.23674	-1.47502	-0.64604	-1.66893	-1.18432	-0.93776	.	.	8.333	Low
11	01JAN2024	UK	5.9	10.8	11.4	12.8	35.3	01JAN1990	34	01JAN2024	Europe	High	-0.05106	-0.97047	-0.53611	1.16472	-0.43781	-0.23423	.	.	50.000	High
12	01JAN2024	USA	14.5	12.3	22.4	33.9	40.2	01JAN1990	34	01JAN2024	North America	High	1.54576	-0.76592	0.10028	0.10084	-0.16277	0.29115	.	.	75.000	Severe
13	01JAN2024	Fiji	1.7	26.6	40.2	50.0	45.0	.	.	.			.	.	.	.	.	.	.	.	.

9) Use PROC SQL to create a summary table by Region & Income_Group

proc sql;

create table region_summary as

select Region, Income_Group,

count(*) as N,

mean(CO2_Emissions) as Mean_CO2 format=8.2,

mean(Average_Temperature) as Mean_Temp format=8.2,

mean(Sea_Level_Rise) as Mean_SL format=8.2,

mean(Forest_Cover) as Mean_Forest format=8.2,

mean(Pollution_Index) as Mean_Poll format=8.2

from climate_final

group by Region, Income_Group

order by Region;

quit;

proc print data=region_summary;

run;

OUTPUT:


Obs	Region	Income_Group	N	Mean_CO2	Mean_Temp	Mean_SL	Mean_Forest	Mean_Poll
1			1	1.70	26.60	40.20	50.00	45.00
2	Africa	Low	1	0.30	19.80	12.00	45.00	34.00
3	Asia	High	1	9.20	13.60	19.10	67.00	29.80
4	Asia	Low	1	0.50	25.10	35.60	17.00	78.90
5	Asia	Lower-Midd	1	1.90	26.80	18.20	24.00	65.50
6	Asia	Upper-Midd	2	4.80	21.25	43.45	27.15	58.50
7	Europe	High	3	6.40	9.47	9.90	37.90	28.60
8	North America	High	1	14.50	12.30	22.40	33.90	40.20
9	Oceania	High	1	16.80	21.50	14.00	16.50	36.00
10	South America	Upper-Midd	1	2.10	25.00	10.10	59.40	30.00

10) Correlation matrix to explore relationships

proc corr data=climate_final nosimple plots=matrix(histogram);

var CO2_Emissions Average_Temperature Sea_Level_Rise Forest_Cover Pollution_Index;

run;

OUTPUT:

The CORR Procedure


5 Variables:	CO2_Emissions Average_Temperature Sea_Level_Rise Forest_Cover Pollution_Index


Pearson Correlation Coefficients, N = 13 Prob > \|r\| under H0: Rho=0
	CO2_Emissions	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index
CO2_Emissions	1.00000	-0.47774 0.0987	-0.31052 0.3018	-0.18307 0.5494	-0.37300 0.2094
Average_Temperature	-0.47774 0.0987	1.00000	0.59813 0.0308	-0.16870 0.5817	0.62744 0.0217
Sea_Level_Rise	-0.31052 0.3018	0.59813 0.0308	1.00000	-0.12244 0.6903	0.60689 0.0278
Forest_Cover	-0.18307 0.5494	-0.16870 0.5817	-0.12244 0.6903	1.00000	-0.57290 0.0407
Pollution_Index	-0.37300 0.2094	0.62744 0.0217	0.60689 0.0278	-0.57290 0.0407	1.00000

11) Regression: model CO2_Emissions as a function of other indicators

proc reg data=climate_final;

model CO2_Emissions = Average_Temperature Sea_Level_Rise Forest_Cover Pollution_Index / vif collin;

output out=reg_out p=Predicted_CO2 r=Residual;

run;

quit;

OUTPUT:

The REG Procedure

Model: MODEL1

Dependent Variable: CO2_Emissions


Number of Observations Read	13
Number of Observations Used	13


Analysis of Variance
Source	DF	Sum of Squares	Mean Square	F Value	Pr > F
Model	4	141.33682	35.33421	1.44	0.3054
Error	8	196.21087	24.52636
Corrected Total	12	337.54769


Root MSE	4.95241	R-Square	0.4187
Dependent Mean	5.83077	Adj R-Sq	0.1281
Coeff Var	84.93578


Parameter Estimates
Variable	DF	Parameter Estimate	Standard Error	t Value	Pr > \|t\|	Variance Inflation
Intercept	1	22.53190	7.54545	2.99	0.0174	0
Average_Temperature	1	-0.20253	0.26974	-0.75	0.4743	1.96136
Sea_Level_Rise	1	0.05058	0.11457	0.44	0.6705	1.94720
Forest_Cover	1	-0.15478	0.09790	-1.58	0.1525	1.76263
Pollution_Index	1	-0.19271	0.15156	-1.27	0.2393	3.27311


Collinearity Diagnostics
Number	Eigenvalue	Condition Index	Proportion of Variation
Number	Eigenvalue	Condition Index	Intercept	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index
1	4.40533	1.00000	0.00162	0.00321	0.00780	0.00454	0.00184
2	0.37187	3.44187	0.00506	0.00274	0.15352	0.16926	0.00754
3	0.15516	5.32847	0.02216	0.02150	0.56155	0.12402	0.05199
4	0.04984	9.40191	0.10629	0.92534	0.08884	0.00261	0.08517
5	0.01781	15.72816	0.86488	0.04722	0.18830	0.69957	0.85348

The REG Procedure

Model: MODEL1

Dependent Variable: CO2_Emissions

Panel of fit diagnostics for CO2_Emissions.

Panel of scatterplots of residuals by regressors for CO2_Emissions.

12) Visualization: scatter matrix and a scatter + regression line for CO2 vs Temp

proc sgscatter data=climate_final;

matrix CO2_Emissions Average_Temperature Sea_Level_Rise Forest_Cover Pollution_Index

/ diagonal=(histogram);

run;

OUTPUT:

/* Scatter + fitted line */

proc sgplot data=reg_out;

title "CO2 Emissions vs Average Temperature with Regression Line";

scatter x=Average_Temperature y=CO2_Emissions / datalabel=Country;

reg x=Average_Temperature y=CO2_Emissions / CLI CLM;

xaxis label="Average Temperature (°C)";

yaxis label="CO2 Emissions (metric tons per capita)";

run;

title;

OUTPUT:

13) Final dataset view

proc print data=climate_final label noobs;

var Country Region Income_Group Observation_Date CO2_Emissions Average_Temperature

Sea_Level_Rise Forest_Cover Pollution_Index Risk_Score Risk_Pct Risk_Category;

label CO2_Emissions = "CO2 (t per capita)";

run;

OUTPUT:


Country	Region	Income_Group	Observation_Date	CO2 (t per capita)	Average_Temperature	Sea_Level_Rise	Forest_Cover	Pollution_Index	Risk_Score	Rank for Variable Risk_Score	Risk_Category
Australia	Oceania	High	01JAN2024	16.8	21.5	14.0	16.5	36.0	0.74374	91.667	Severe
Bangladesh	Asia	Low	01JAN2024	0.5	25.1	35.6	17.0	78.9	0.44545	83.333	Severe
Brazil	South America	Upper-Midd	01JAN2024	2.1	25.0	10.1	59.4	30.0	-0.35904	25.000	Moderate
China	Asia	Upper-Midd	01JAN2024	7.6	14.2	16.7	22.3	55.0	0.07642	58.333	High
Ethiopia	Africa	Low	01JAN2024	0.3	19.8	12.0	45.0	34.0	-0.48323	16.667	Low
Germany	Europe	High	01JAN2024	8.4	10.5	8.8	31.9	28.5	-0.31791	33.333	Moderate
India	Asia	Lower-Midd	01JAN2024	1.9	26.8	18.2	24.0	65.5	0.25190	66.667	High
Japan	Asia	High	01JAN2024	9.2	13.6	19.1	67.0	29.8	-0.30665	41.667	Moderate
Maldives	Asia	Upper-Midd	01JAN2024	2.0	28.3	70.2	32.0	62.0	0.83014	100.000	Severe
Sweden	Europe	High	01JAN2024	4.9	7.1	9.5	69.0	22.0	-0.93776	8.333	Low
UK	Europe	High	01JAN2024	5.9	10.8	11.4	12.8	35.3	-0.23423	50.000	High
USA	North America	High	01JAN2024	14.5	12.3	22.4	33.9	40.2	0.29115	75.000	Severe
Fiji			01JAN2024	1.7	26.6	40.2	50.0	45.0	.	.

To Visit My Previous Original Data Management Dataset:Click Here

To Visit My Previous Cricket World Cup Dataset:Click Here

To Visit My Previous Home Tour Dataset:Click Here

To Visit My Previous Real-World Dal Price Dataset:Click Here

Follow Us On :

--- FOLLOW OUR BLOG FOR MORE INFORMATION.

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

Search This Blog

SAS Learning Hub

326.How Fast Is Our Planet Warming? A Data-Driven Climate Change Study in SAS

How Fast Is Our Planet Warming? A Data-Driven Climate Change Study in SAS

Follow Us On :

Comments

Post a Comment

Popular posts from this blog

409.Can We Build a Reliable Emergency Services Analytics & Fraud Detection System in SAS While Identifying and Fixing Intentional Errors?

397.If a satellite has excellent signal strength but very high latency, can it still deliver good quality communication? Why or why not?A Sas Study

401.How Efficient Are Global Data Centers? A Complete SAS Analytics Study