Agricultural eco-efficiency and climate determinants: application of dea with bootstrap methods in the tropical montane cloud forests of Puebla, Mexico

Figures & data

Figure 1. Municipalities of Puebla, located in the TMCF.

Table 1. Descriptive statistics of inputs (x), outputs (y), and environmental determinants

	x1	x2	x3	y1	y2	T	P
Average	2124.44	5667.62	332.98	47171578.94	2694.39	17.10	2316
Median	1545.00	3747.17	250.00	17190019.53	2499.48	16.96	2234
Standard deviation	1702.94	5011.00	284.70	76477785.67	2107.35	1.548	565.8
Minimum	50.00	109.95	86.00	359510.05	60.55	14.43	1212
1st quartile	851.80	1578.60	167.50	9294241.00	1253.19	15.99	1977
3rd quartile	3073.50	8649.20	356.00	60205400.00	3418.06	18.14	2610
Maximum	6078.00	22458.00	1421.00	505135644.70	8787.10	21.89	4098

Table 2. Matrix of outliers for the initial sample

r	Deleted observations	${\mathrm{R}}_{min}^{\left(\mathrm{r}\right)}$
1	45	0.2015
2	54 45	0.1353
3	53 54 45	0.0772
4	23,535,445	0.0517
5	2,324,535,445	0.0266
6	232,453,185,445	0.0174
7	23,245,318,175,445	0.0095
8	2,324,555,318,175,445	0.0063

Graph 1. Log-ratio for the initial sample.

Graph 2. Bias-corrected efficiency estimates and 95% confidence intervals (Source: elaborated by the authors).

Table 3. Eco-efficiency

Dmu	Eco-efficiency without bias correction	Eco-efficiency with bias correction	95% confidence interval
			Lower limit	Upper limit
HUT16	2.028	1.094	1.009	1.207
CUT16	1.039	1.097	1.028	1.206
YAN16	1.000	1.114	1.009	1.264
. . .
AHC16	3.091	2.789	2.624	3.090
AHC17	2.942	2.830	2.644	3.128
TLT17	3.525	3.833	3.433	4.548
AVERAGE (STANDARD DEVIATION)	2.40 (2.44)	1.59 (0.58)	1.44 (0.46)	1.82 (0.64)

Source: Elaborated by the authors.

Graph 3. Boxplot of the corrected eco-efficiency indexes for the groups (Source: elaborated by the authors).

Table 4. Corrected eco-efficiency statistics for the Kolmogorov-Smirnov and Kruskal-Wallis groups and tests

	Group 1—municipalities 2016	Group 2—municipalities 2017
Minimum	1.094	1.134
1st Q.	1.131	1.173
Median	1.259	1.259
Average	1.546	1.627
3rd Q.	1.981	1.984
Maximum	2.789	3.833

Two-sample Kolmogorov-Smirnov test: d = 0.29032, p-value = 0.1151Kruskal-Wallis chi-squared test = 1.1207, df = 1, p-value = 0.2898Source: Elaborated by the authors

Figure 2. Temperature, precipitation, and eco-efficiency of 2016 in the study area.

Figure 3. Temperature, precipitation, and eco-efficiency of 2017 in the study area.

Table 5. Second stage model estimates

Model	Coefficient	95% confidence interval
		Lower limit	Upper limit
Intercepto	−36.4667100	−38.27547000	−34.65794000
T	4.9515190	4.72098800	5.18204900
P	−0.0014957	−0.00165565	−0.00133573
T^2	−0.1497322	−0.15647300	−0.1429913
P^2	0.0000005	0.00000043	0.0000005
Error deviation	1.4469570	1.44072400	1.45319000

Average marginal effect of temperature = −0.1610705Average Marginal Effect of Precipitation = 0.00065Source: Elaborated by the authors

Data Availability Statement

The database used and the data sources can be found in the open access link located in: https://data.mendeley.com/datasets/gf6xnkb5tw/1 and doi:10.17632/gf6xnkb5tw.1.