Received: 12/08/2024
Accepted: 09/09/2024
https://revistas.unj.edu.pe/index.php/pakamuros
126
Volumen 12, Número 3, Julio-Septiembre, 2024, Páginas 126 al 136
DOI: https://doi.org/10.37787/s7ztng79
ORIGINAL ARTICLE
Pronóstico de la competitividad exportadora del café peruano en los principales países
importadores utilizando el modelo autorregresivo integrado de media móvil
Forecasting the export competitiveness of peruvian coffee in major importing countries using the
integrated autoregressive moving average model
Roger Abanto
1
*, Flor Cabrera
2
, Jose Ruiz1 y Juan Rodriguez1
RESUMEN
Este estudio examina la competitividad de las exportaciones de café peruano utilizando el modelo autorregresivo
integrado de medias móviles (ARIMA) para predecir las tendencias de los principales países importadores.
Mediante el análisis de series temporales de datos sobre volúmenes y precios de exportación, el modelo ARIMA
demostró su eficacia en la predicción de tendencias futuras, proporcionando información valiosa sobre los factores
que influyen en la rentabilidad y sostenibilidad del sector. Los resultados revelan que el mercado peruano del café
está sujeto a importantes fluctuaciones, influidas por la volatilidad de los precios mundiales y la competencia.
Además, el estudio subraya el papel fundamental de la mejora de las infraestructuras y el apoyo gubernamental
para impulsar la competitividad del sector, existen también mejoras internas, como la optimización de los procesos
y la adopción de tecnología, para que los productores peruanos de café mantengan su posición en el mercado. En
consonancia con la literatura, el estudio confirma que las economías de escala son fundamentales para reducir los
costes de producción y mejorar la eficiencia. Aprovechando modelos predictivos como ARIMA, los productores
y los responsables políticos pueden tomar decisiones estratégicas informadas, asegurando la competitividad a largo
plazo del café peruano en el mercado mundial.
Palabras clave: Competitividad, ARIMA, exportación, café.
ABSTRACT
This study examines the competitiveness of Peruvian coffee exports using the autoregressive integrated moving
average model (ARIMA) to predict trends in major importing countries. By analyzing time series data on export
volumes and prices, the ARIMA model proved to be effective in predicting future trends, providing valuable
information on the factors that influence the profitability and sustainability of the sector. The results reveal that
the Peruvian coffee market is subject to significant fluctuations, influenced by world price volatility and
competition. In addition, the study highlights the critical role of improved infrastructure and government support
in boosting the competitiveness of the sector, there are also internal improvements, such as process optimization
and technology adoption, for Peruvian coffee producers to maintain their position in the market. In line with the
literature, the study confirms that economies of scale are key to reducing production costs and improving
efficiency. By leveraging predictive models such as ARIMA, producers and policy makers can make informed
strategic decisions, ensuring the long-term competitiveness of Peruvian coffee in the world market.
Keywords: Competitiveness, ARIMA, exportation, coffee.
* Corresponding author
1
National University of Trujillo, Peru. Email: rabantod@unitru.edu.pe, jsirlopu@unitru.edu.pe,
jrodriguezmon@unitru.edu.pe
2
Señor de Sipan University, Perú. Email: csernaqueflorro@uss.edu.pe
Abanto et al.
127
INTRODUCTION
The Peruvian coffee sector has played a crucial role in the country's economy, standing out as one of its
main export products (Valdiglesias, 2024) and is recognized for its diverse ecosystems. It offers a wide
variety of high-quality coffees, including the renowned Arabica and Robusta varieties (Dilas et al., 2021).
Coffee-producing regions such as Cajamarca, Junín, and San Martín are famous for their beans, which
possess unique flavor profiles that are highly valued in international markets (Dilas & Cernaqué, 2017;
Rojas et al., 2021).
The export of Peruvian coffee has shown consistent growth, contributing significantly to the national
economy (Díaz et al., 2018). However, the sector faces challenges related to international price volatility
and global competition (Rivera, 2022). Factors such as production costs, supply chain efficiency, and the
producers' ability to achieve economies of scale play a crucial role in the sector's competitiveness
(Montes & Oblitas, 2023). Operational improvements and industry evolution are essential to understand
how coffee producers in Peru can increase efficiency and minimize costs as they expand their production
and exports (Cerquera et al., 2020). Internal improvements focus on benefits that each company gains by
increasing production volume (Figueroa et al., 2019), such as process optimization and adopting new
technologies. On the other hand, external improvements refer to advancements across the entire Peruvian
coffee industry, which may include developing better infrastructure, such as roads and processing
technology (Vera et al., 2024), as well as implementing government policies that support the sector's
growth and competitiveness at the global level. These improvements benefit not only large companies
(Sacco et al., 2011), but also help small producers better integrate into international markets.
In this study, we analyze the impact of export competitiveness in the Peruvian coffee sector, focusing on
export volumes and prices (Ortiz et al., 2004), using the Autoregressive Integrated Moving Average
method (ARIMA). Utilizing this methodology, we will model and evaluate export trends, providing a
deeper understanding of the factors influencing the profitability and sustainability of this vital sector for
the Peruvian economy. The ARIMA model is particularly suitable for this analysis due to its ability to
model and predict time series data that may show trends and seasonal patterns (Sánchez et al., 2013).
This methodology allows us to identify demand dynamics in major importing markets and predict future
price and export volume trends (González, 2009). The accuracy of ARIMA in the context of coffee trade
is due to its integrated approach that considers both historical fluctuations and seasonal variations
(Camones, 2022), which is crucial for developing effective export strategies. This analysis provides
Forecasting the export competitiveness of Peruvian coffee
128
producers and policymakers with valuable insights to make informed and strategic decisions aimed at
maximizing competitiveness and reach in international markets (Aguilar, 2022).
MATERIALS AND METHODS
The ARIMA model is a fundamental tool in time series analysis, widely used in statistics and
econometrics to predict future data from historical values. Defined as ARIMA (p, d, q), the model
combines autoregressive terms (p), differentiations (d) to achieve stationarity, and moving average terms
(q) to adjust prediction errors. This flexible structure allows ARIMA to capture temporal dependencies
and underlying patterns in data, offering a robust and adaptable representation applicable in fields as
diverse as economics and engineering (Chávez, 1997).
Autoregression - AR(p)
A part of the ARIMA model is the autoregressive model of order p (AR(p)), which uses dependencies
between successive observations. The equation is:
Where:
ϕ1, ϕ2…, ϕp are the parameters of the model
c is a constant
ϵt is the error term at time t
Integration - I(d)
The integration (I) of order d in ARIMA indicates that the time series has been differenced d times to
achieve stationarity.
B is the delay operator
Moving Average - MA(q)
The moving average of order q (MA(q)) models the error term as a linear combination of error terms in
past observations:
θ1, θ2…, θq are the parameters of the moving average
ηt is the white noise
Full ARIMA Model (p, d, q)
Dickey-Fuller test to verify stationarity and determine d.
Abanto et al.
129
ACF and PACF plots to determine p and q.
AIC/BIC-based model selection to fit the most appropriate model.
RESULTS
Table 1
World trend of the main coffee importing countries 2004 - 2023 (greater than 2k Tons) main exporting countries of
unroasted and decaffeinated coffee exported by Peru.
Item
World
Alemania
USA
Belgium
Colombia
Sweden
Canada
Korea
Italy
France
UK
Netherlands
Japan
Spain
2004
191131
67315
48978
22458
700
3917
5794
3179
1633
6816
5443
10258
2530
1207
2005
142151
46470
32780
7522
3858
5535
4782
3545
2200
6300
4675
10866
2596
2149
2006
238063
83468
50972
32921
5715
6750
5381
4848
2677
4513
4371
13745
7873
2240
2007
173615
54868
52739
15504
1037
7869
5321
5609
2436
2606
4382
7886
2208
2443
2008
224648
73091
55978
34005
3937
9275
5023
8031
6330
1860
4573
2821
3073
2410
2009
197470
61952
43214
23770
20285
7048
3986
5913
4548
3182
3093
2393
2642
1347
2010
229617
81226
48748
23235
13683
9052
7244
8351
6820
4838
5045
2031
3029
1962
2011
296348
84920
65235
48368
24974
8121
8106
9051
6453
3940
5346
4861
3117
2048
2012
266288
89018
43690
35873
30374
9366
8854
8376
6411
5633
4449
4080
1780
1188
2013
238645
82439
52162
23627
13861
11032
6270
11007
6749
6497
5032
2733
3279
1347
2014
185138
50885
44246
20582
6915
9217
6268
11651
6415
5243
4581
2760
1376
1012
2015
174999
50360
42575
19017
2182
9430
8635
10299
3606
5331
4519
2428
1756
1122
2016
239331
59062
65014
23488
8834
11349
9463
9030
7957
7084
6165
4832
1585
1809
2017
245735
54543
58784
22847
16253
14004
11535
10153
10241
4065
6862
5683
2951
2890
2018
256272
57064
62971
28410
23232
14201
11504
7005
8824
6900
8422
4764
4637
1473
2019
226540
50424
57995
21419
11877
13070
11449
8546
8861
7501
6977
5469
4918
1947
2020
213215
46749
53113
18445
17780
10700
9502
9095
9680
10746
5218
1972
4312
2285
2021
191600
38748
42863
20294
22066
9055
9601
6421
8378
8013
6574
2895
2459
2266
2022
246471
50086
54478
29696
23503
12447
10954
6774
8823
6571
8963
7512
5607
3675
2023
204547
37625
56025
18408
12690
8766
12150
4710
8764
7554
6667
8679
4001
2708
Note: Fluctuations are observed with peaks as in 2011, when exceptionally high volumes were recorded globally, especially in Germany,
the United States and Belgium. Regional variability is evident, with some countries maintaining consistently high volumes, while others,
such as Korea and Spain, show more modest but fluctuating volumes. These variations can be influenced by external factors such as trade
policies, coffee production and quality, and global economic conditions, thus offering a comprehensive view of the dynamics of the global
coffee market. TRADEMAP, (2024).
Figure 1
World trend of the main coffee importing countries 2004 - 2023 (greater than 2k Tons) main exporting countries of
unroasted and decaffeinated coffee exported from Peru
Forecasting the export competitiveness of Peruvian coffee
130
Table 2
World trend of the main coffee importing countries 2004 - 2023 (greater than 3K thousand USD) main exporting countries
of unroasted and decaffeinated coffee exported by Peru
Item
World
USA
Germany
Belgium
Canada
Sweden
Netherlands
Italy
Francia
UK
Colombia
Korea
Japan
Spain
2004
289844
74902
98755
34788
9653
6336
17088
2533
10411
9887
490
4797
4111
1691
2005
306075
71429
98335
17549
10915
12709
24771
4572
13130
11155
4942
7428
5853
3976
2006
514918
112257
181239
72900
12869
15518
31368
5803
9953
10964
7449
10137
12593
4544
2007
426884
125259
134614
39070
14532
20973
19757
6082
6548
12183
1365
13673
5767
5167
2008
643800
158008
211851
100560
15943
28407
8093
18210
5333
14359
4991
21696
9062
6486
2009
583784
190183
128058
75195
23375
38822
13250
19075
13783
10303
9850
7307
8095
3886
2010
887045
314029
190216
96780
37874
32582
29683
35476
26502
21067
18948
8689
11836
7554
2011
1596751
470985
372175
271974
49248
74707
48270
50784
35921
32158
22133
27157
18421
11198
2012
1022848
346328
187425
136966
39123
85708
36095
30865
25112
19086
22086
15105
7377
4288
2013
698758
243740
156695
69200
36162
23005
21756
30960
20222
17721
19109
8847
9719
3704
2014
747838
214517
178909
86476
40829
8706
29187
47792
24721
21345
22451
11744
6475
3127
2015
579586
163961
147435
65873
33463
2939
33253
31996
11081
16553
16491
7937
6282
2917
2016
756333
193111
216800
79953
40430
9365
35233
26781
24505
22805
22468
16730
5646
5226
2017
708822
167837
182040
72072
44142
18538
37922
29330
27340
22329
12735
17629
9424
6341
2018
667336
153095
178760
77170
39825
32115
33604
16390
22139
24820
17572
13047
12770
3642
2019
619656
141538
169337
60012
37336
16474
34123
20914
22477
21509
19938
16250
13057
4893
2020
639890
141541
172836
58358
33941
26947
31909
26110
30537
17176
32425
6217
13711
5726
2021
756687
163476
177479
87517
39407
45993
37614
25218
36589
28610
32834
12827
11028
7985
2022
1234294
259079
283605
154285
60115
78183
55951
35808
45775
47042
33340
42267
31684
16854
2023
827514
150785
230096
76596
34876
28294
58917
21575
34422
29392
31242
34704
17813
10894
Note: The table shows coffee imports in thousands of USD from 2004 to 2023. Globally, there are notable fluctuations, with a peak in
2011. The United States and Germany are the largest importers, followed by Belgium, Canada, and Japan. The table reflects the economic
dynamics of the global coffee market, influenced by demand and prices. TRADEMAP, (2024).
Figure 2
World trend of the main coffee importing countries 2004 - 2023 (greater than 3K thousand USD) main exporting countries
of unroasted and non-decaffeinated coffee exported by Peru
Note: The graph shows coffee imports with values exceeding 3,000 USD from 2004 to 2023. The United States and Germany are the largest
importers, with a notable peak in 2011. The red bars represent the global total, and the colored lines represent the imports of each country.
This trend reflects the dynamics of the global coffee market. TRADEMAP, (2024).
0
200000
400000
600000
800000
1000000
1200000
1400000
1600000
1800000
0
50000
100000
150000
200000
250000
300000
350000
400000
450000
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
Wolrd
USA
Germany
Belgium
Canada
Sweden
Netherlands
Italy
France
UK
Colombia
Korea
Japan
Spain
Abanto et al.
131
Dickey-Fuller test to evaluate stationarity:
If the p value of the ADF test is less than 0.05, the series is stationary and does not need further
differentiation (d=0).
If the p-value of the ADF test is greater than 0.05, the series is not stationary and needs differentiation
(d=1).
Table 3
Values (d,p,q) for each country and Dickey-Fuller test
Country
ADF Statistic
p-value
AIC
BIC
Germany
-2.24391
0.191
422.542
425.376
USA
-3.86932
0.002
431.068
434.055
Belgium
-4.19348
0.001
426.646
430.629
Colombia
-4.20763
0.001
423.149
426.136
Sweden
-2.52379
0.110
347.331
351.108
Canada
-0.0012081
0.958
337.133
340.910
Korea
-1.63088
0.467
343.102
346.880
Italy
-1.03998
0.738
339.144
342.922
France
-1.03998
0.738
344.812
348.590
UK
-0.737904
0.837
331.117
334.895
Netherlands
-0.682280
0.851
356.523
360.301
Japan
-3.97765
0.002
358.422
362.405
Spain
-2.45145
0.128
308.523
312.301
Note: Stationarity: Countries such as the USA, Belgium, Colombia, and Japan have significantly low p-values (< 0.05) and sufficiently
large negative ADF statistics, suggesting that their series are stationary without the need for differencing. This implies that the time series
for these countries do not show consistent trends or seasonality that vary over time.
Non-Stationary: Canada, South Korea, Italy, France, the United Kingdom, the Netherlands, Spain, and Sweden exhibit relatively high p-
values, indicating that we cannot reject the null hypothesis of a unit root, suggesting that these series need differencing to be considered
stationary. Additionally, the use of AIC and BIC is crucial for selecting the most appropriate model. In the case of Sweden and Canada,
where AIC and BIC are relatively low compared to other countries with d=1d=1d=1, it could indicate that the differencing process was
effective and/or that the fitted model is relatively simple and adequate for those data.
ARIMA (d,p,q) (0,1,1)
Developing the model
This can be interpreted as follows:
Xt is the current value of the time series.
Xt-1 is the previous value of the time series.
ϵt is the white error at time t.
θ1 is the parameter of the first order moving average term.
for the ARIMA (0,1,1) model, the full model equation is:
This shows that the differenced value of the series at time t depends on the error at time t and the error
at time t-1.
ARIMA (d,p,q) (1,1,1)
Forecasting the export competitiveness of Peruvian coffee
132
Expanding this equation, we first expand the left side:
Simplifying:
In such a way that:
Xt The current value of the time series.
Xt-1 It is the previous value of the time series.
Xt-2 It is the value from two previous periods in the time series.
ϵt It is the white noise error at time t.
ϵt-1 It is the white noise error at time t-1
ϕ1 It is the parameter of the first-order autoregressive term.
θ1 It is the parameter of the first-order moving average term.
For the ARIMA (1,1,1) model, the complete equation of the model is:
Table 4
Unit Root Analysis and ARIMA Parameters by Country
Country
ADF Statistic
p-value
d
p
q
AIC
BIC
Germany
-2.24391
0.191
1
1
1
422.542
425.376
USA
-3.86932
0.002
0
1
1
431.068
434.055
Belgium
-4.19348
0.001
0
1
1
426.646
430.629
Colombia
-4.20763
0.001
0
1
1
423.149
426.136
Sweden
-2.52379
0.110
1
1
1
347.331
351.108
Canada
-0.0012081
0.958
1
1
1
337.133
340.910
Corea
-1.63088
0.467
1
1
1
343.102
346.880
Italy
-1.03998
0.738
1
1
1
339.144
342.922
France
-1.03998
0.738
1
1
1
344.812
348.590
UK
-0.737904
0.837
1
1
1
331.117
334.895
Netherlands
-0.682280
0.851
1
1
1
356.523
360.301
Japan
-3.97765
0.002
0
1
1
358.422
362.405
Spain
-2.45145
0.128
1
1
1
308.523
312.301
Note: The table shows the results of the Dickey-Fuller test and the ARIMA parameters for coffee imports. The USA, Belgium,
Colombia, and Japan have stationary series (p<0.05, d=0), while other countries require differencing (d=1). The parameters p
and q, along with the AIC and BIC values, help identify the best ARIMA models for each country.
Abanto et al.
133
Table 5
ARIMA Model Parameters and Component Analysis Results by Country
Country
d
p
q
Autoregressive
Integration
Moving Average
Constant C
Coefficient
AR (1) ϕ
I (1); I (0);
MA (1)
Germany
1
1
1
30302.05
0.488091
I (1)
0.247569
USA
0
1
1
48753.97
0.058631
I (0)
0.065468
Belgium
0
1
1
25326.99
-0.029231
I (0)
-0.023915
Colombia
0
1
1
6997.58
0.518198
I (1)
0.552572
Sweden
1
1
1
3799.58
0.628836
I (1)
0.753522
Canada
1
1
1
1193.99
0.890895
I (1)
0.711776
Korea
1
1
1
2611.84
0.672571
I (1)
0.690679
Italy
1
1
1
2202.88
0.708306
I (1)
0.570161
France
1
1
1
2093.32
0.637351
I (1)
0.461701
UK
1
1
1
2190.59
0.614126
I (1)
0.667981
Netherlands
1
1
1
1674.09
0.666096
I (1)
0.999957
Japan
0
1
1
3207.36
0.036598
I (0)
0.030575
Spain
1
1
1
1264.60
0.388200
I (1)
0.440389
Note: The table presents the parameters of the ARIMA model (p, d, q) and the autoregressive (AR), integration (I), and moving
average (MA) components for coffee imports in various countries. The countries with stationary series without differencing
(d=0) include the USA, Belgium, Colombia, and Japan, while others require differencing (d=1). The constant C and the AR
coefficient (ϕ) vary by country, indicating the relationship between the current value and the past value of imports. The MA
coefficients (θ) reflect the influence of past errors.
Table 6
ARIMA Models and Autoregressive Components by country
Country
Model
Germany
;
USA
;
Belgium
;
Colombia
;
Sweden
;
Canada
;
Korea
;
Italia
;
France
;
UK
;
Netherlands
;
Japan
;
Spain
;
Note: The table presents the ARIMA (p, d, q) models specific to coffee imports in various countries. Each model is described
by an autoregressive equation with a constant, an AR (ϕ) coefficient, and a moving average (MA) term. These models reflect
the relationship between the current value of imports and their past values, adjusted by an error term that considers the
influence of past errors. Each country has a specific model that captures the particular dynamics of its coffee imports.
Forecasting the export competitiveness of Peruvian coffee
134
Table 7
Projection of Imports by Country (2024-2030)
Year
2024
2025
2026
2027
2028
2029
2030
Germany
48666.47
66104.03
78932.14
88369.27
95311.79
100419.12
104176.38
USA
52038.77
55211.93
55605.72
55654.58
55660.65
55661.40
55661.49
Belgium
24788.91
24009.56
24050.98
24048.78
24048.89
24048.89
24048.89
Colombia
13573.51
21531.69
30053.07
39177.5
48947.68
59409.28
70611.26
Sweden
9311.96
16672.04
26846.3
40910.78
60352.93
87228.93
124381.19
Canada
12018.36
20455.47
33977.38
55648.56
90380.32
146043.91
235254.33
Corea
5779.65
10490.95
16913.62
25669.34
37605.56
53877.63
76060.51
Italia
8410.47
12955.39
18765.92
26194.49
35691.67
47833.51
63356.44
Francia
6907.87
9685.43
12738.11
16093.16
19780.54
23833.17
28287.21
UK
6284.97
10248.59
15330.38
21845.78
30199.22
40909.22
54640.58
Netherlands
7455.14
14094.74
25156.68
43586.45
74291.43
125447.55
210676.35
Japan
3353.79
3432.64
3437.94
3438.3
3438.32
3438.32
3438.32
Spain
2315.85
3183.48
3902.4
4498.09
4991.66
5400.64
5739.51
Note: The table shows coffee import forecasts (in thousands of USD) for 2024-2030. Germany and the USA maintain stable
trends, while Colombia, Sweden, and Canada experience significant growth. Japan and Spain show stability in their imports.
These forecasts reflect the expectations in the global coffee market.
DISCUSSION
The analysis of the export competitiveness of Peruvian coffee using the ARIMA model allowed us to
identify patterns and trends in export volumes. In contrast to Rivera (2022), who highlights the stability
of Mexican coffee competitiveness amidst global variability, our results indicate that the Peruvian coffee
market exhibits significant fluctuations. This underscores the importance of differencing in time series
to achieve stationarity, a crucial aspect also supported by Camones (2022) in the context of coffee trade.
Additionally, the study by Cerquera et al. (2020) on infrastructure and governmental policies emphasizes
the need for external improvements for the growth of the coffee sector. Our findings align with this
perspective, highlighting the relevance of infrastructure enhancement and government support as key
elements to increase the competitiveness of Peruvian coffee in the international market. Furthermore, the
works of Figueroa et al. (2019) and Sacco et al. (2011) suggest that optimizing internal processes and
adopting new technologies can improve efficiency and reduce costs in coffee production. Our analysis
supports this claim, demonstrating that the ability of Peruvian producers to adapt to these internal
improvements is essential for maintaining their competitiveness. Consistent with the results of Montes
and Oblitas (2023), we find that economies of scale play a decisive role in the sector's competitiveness,
emphasizing the need to increase production volume to optimize costs. The implementation of export
strategies informed by predictive models, such as ARIMA, as suggested by Sánchez et al. (2014, is
fundamental for anticipating trends and adjusting export strategies. Our results confirm the effectiveness
of the ARIMA model for this purpose, providing producers and policymakers with valuable insights for
Abanto et al.
135
making strategic decisions aimed at maximizing competitiveness and market reach in international
markets.
CONCLUSIONS
The ARIMA model has proven to be highly effective in forecasting coffee import trends in major
importing countries. Its ability to capture complex patterns in time series and differentiate non-stationary
series has been crucial for obtaining accurate and reliable forecasts, providing a solid foundation for
strategic decision-making in the coffee sector.
The results show that the ARIMA model is versatile, allowing for forecasts without the need for
differencing in countries with stationary series and adapting through differencing in those with non-
stationary series. This highlights its applicability in various market contexts and historical data.
The application of the ARIMA model has provided valuable insights for producers and policymakers,
enabling them to anticipate fluctuations in demand and prices. This facilitates the creation of informed
and effective strategies, contributing to maintaining the competitiveness of Peruvian coffee in the global
market. Continuous monitoring and adjustment of forecasting models are essential to ensure strategic
decisions based on accurate and up-to-date data.
BIBLIOGRAPHICAL REFERENCES
Aguilar, B. A., & Acosta Abreu, R. S. (2022). Ajuste de un modelo ARIMA para los precios de exportación del aguacate
calibre 60. México: Escuela Superior de Física y Matemáticas . Retrieved from
https://www.esfm.ipn.mx/assets/files/esfm/docs/RNAFM/articulos-2022/XXVIIRNAFM002.pdf
Camones Gonzales, F. C. (2022). Control de calidad en los procesos estadísticos: Una aproximación basada en los modelos
ARIMA con Análisis de Intervención. INEI. Retrieved from
https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib0516/Libro.pdf
Cerquera Losada, O. H., Pérez Gómez, V. H., & Sierra Chavarro, J. (2020). Análisis de la competitividad de las exportaciones
del café del Huila. Tendencias, Vol. 21 N.° 2, 19-44.
Chávez Quisbet, N. (1997). Modelos ARIMA. Revista Ciencia y Cultura, 23-30.
Díaz Rodríguez, A., Silva Jaimes, M. I., & Dávila Romero, J. (2018). Relación entre las buenas prácticas de higiene y la
ocurrencia de ocratoxina A en café (Coffea arabica L.) orgánico de las principales zonas cafetaleras del Perú.
Scientia Agropecuaria, 11. doi:http://dx.doi.org/10.17268/sci.agropecu.2018.02.02
Dilas Jiménez, J. O., Diaz Dumont, J. R., Torres Huamaní, J., Mendoza Pumapillo, J. E., & Tamayo Ortiz, C. V. (2021).
Producción, precios y dinámica de las exportaciones del café peruano. Revista de investigación científica y
tenológica Llamkasun, Vol. 2 N.° 2, 12. doi:https://doi.org/10.47797/llamkasun.v2i4.66
Forecasting the export competitiveness of Peruvian coffee
136
Dilas, J., & Cernaqué, O. (2017). El sector cafetalero peruano: Un enfoque a la CTI para su competitividad. Huancayo.
Retrieved from https://hdl.handle.net/20.500.12394/3477
Figueroa Hernández, E., Pérez Soto, F., Godínez Montoya, L., & Perez Figueroa, R. A. (2019). Los precios de café en la
producción y las exportaciones a nivel mundial. Revista Mexicana de Economía y Finanzas Nueva Época, Vol. 14
N.° 1, 41-56. doi:http://dx.doi.org/10.21919/remef.v14i1.358
González Casimiro, M. P. (2009). Análisis de series temporales: Modelos ARIMA. Universidad del País Vasco.
Montes Ninaquispe, J. C., Santa María, A. P., & Oblitas Vallejo, I. A. (2023). Dinámica comercial de las exportaciones
peruanas de café a la Unión Europea 2017-2021. ALFA. Revista de Investigación en Ciencias Agronómicas y
Veterinarias, Vo. 7 N.° 17, 175-183. doi:https://doi.org/10.33996/revistaalfa.v7i19.207
Ortiz Ceballos, G., de la Cruz Vargas Mendoza, M., Mendoza Briseño, M. A., Ojeda Ramírez, M. M., & Trujillo Ortega, L.
(2004). Análisis comparativo de la producción – demanda del café en el Mercado Internacional (1980-2003).
Interciencia, Vol.29 N.° 11 , 621-625. Retrieved from
https://ve.scielo.org/scielo.php?script=sci_arttext&pid=S0378-18442004001100005
Rivera Rojo, C. R. (2022). ompetitividad del café mexicano en el comercio internacional: un análisis comparativo con Brasil,
Colombia y Perú (2000 – 2019). Análisis Económico, Vol. XXXVII, N.° 94, 181-199.
doi:https://doi.org/10.24275/uam/azc/dcsh/ae/2022v37n94/Rivera
Rojas Ruiz, R., Alvarado Huamán, L., Borjas Ventura, R., Carbonell Torres, E., Castro Cepero, V., & Julca Otiniano, A.
(2021). Sustentabilidad en fincas productoras de café (Coffea arabica L.) convencional y orgánica en el Valle del
Alto Mayo, Región San Martín, Perú. RIVAR (Santiago), Vol. 8 N.° 23, 13.
doi:http://dx.doi.org/10.35588/rivar.v8i23.4916
Sacco dos Anjos, F., Belik, W., & Velleda Caldas, N. (2011). La caficultura en Brasil Evolución, situación actual y nuevos
retos cara al futuro. Mundo agrario, Vol. 12 N.° 23, 26.
Sánchez López, E., Barreras Serrano, A., Pérez Linares , C., Figueroa Saavedra, F., & Olivas Valdez, J. A. (2013). Aplicación
de un modelo ARIMA para pronosticar la producción de leche de bovino en baja California. Tropical and
Subtropical Agroecosystems, Vol. 16 N.° 3, 315-324. Retrieved from
http://www.redalyc.org/articulo.oa?id=93929595004
Sánchez, L., Cabanas, G., Abad, Y., & Torres, V. (2014). Utilización de modelos ARIMA para la predicción de la producción
de leche. Estudio de caso en la UBPC “Maniabo”, Las Tunas. Revista Cubana de Ciencia Agrícola, Vol. 48 N.° 3,
213-218. Retrieved from http://www.redalyc.org/articulo.oa?id=193032133002
Valdiglesias Oviedo, J. (2024). Associativity of small farmers and exports in the Peruvian region of Cajamarca. Apuntes, Vol.
51 N° 96. doi:http://dx.doi.org/10.21678/apuntes.96.1970
Vera Velásquez, F. B., Martin Fernández, R. A., & Esquivel García, R. (2024). Diagnóstico de la producción cafetalera en el
cantón Jipijapa, provincia de Manabí, Ecuador. Revista Arbitrada Interdisciplinaria KOINONIA, Año IX. Vol IX.
N°17, 21. doi:http://doi.org/ 10.35381/r.k.v8i17.3146
