"Climate Insurance" package was developed for interested parties to design index-based climate insurance and analyze their climate risk reduction potential. The package requires the user to enter a period, crop yield and index information. Data can be entered in any delineation (eg. space, comma). Consequently, the package designs index-based climate insurance products based on entered data and calculates descriptive statistics, and conducts performance analysis.
- Threat score (aka, critical success index) measures the fraction of yield loss events that were correctly predicted, given by hits/(hits + misses + false alarms)
- Probability of detection (aka, hit rate) measures the probability of an index to detect yield loss events. It is given by hits/(hits + misses).
- False alarm ratio indicates the probability of a yield loss event being detected by mistake. It is given by (false alarms)/(hits + false alarms).
- Pearson’s Correlation Coefficient (PC) measures the linear correlation between index and crop yield
- Hedging effectiveness is the change of expected shortfall of the revenue by the use of an index-based insurance contract. Basically, relative hedging effectiveness it compares the downside risk measure semi-variance of the revenue of uninsured crop yields with insured. Higher relative hedging effectiveness corresponds to lower basis risk and higher risk-reducing effectiveness of the insurance contract. The details about the method can be found in the following publication, Kolle et al. (2021).
- Preference of moderate risk-averse farmers for choosing being insured or uninsured have been calculated the based on expected utility model following the publication by Bucheli et al. (2021).
Current pakcage has been developed for scientific purposes within the project “KlimALEZ – Increasing climate resilience via agricultural insurance – Innovation transfer for sustainable rural development in Central Asia” implemented by Leibniz Institute of Agricultural Development in Transition Economies (IAMO) and funded by German Federal Ministry of Education and Research (BMBF), Germany.
You can install the package using the "devtools":
# install.packages("devtools")
devtools::install_github("iamo-klimalez/climate-insurance")If you use this package in your research, please consider citing it. You can get citation information with
citation("climateinsurance")
#>
#> To cite climateinsurance in publications use:
#>
#>
#>
#> A BibTeX entry for LaTeX users is
#>
#> @Manual{,
#> title = {Improving risk reduction potential of weather index insurance by spatially downscaling gridded climate data - a machine learning approach},
#> author = {Sarvarbek Eltazarov, Ihtiyor Bobojonov, Lena Kuhn, Thomas Glauben},
#> year = {2023},
#> note = {peer-review paper},
#> doi = {https://doi.org/10.1080/20964471.2023.2196830},
#> }In this example we easily design and analyze the performance of the below given index (cummulative precipitation) based insurance products.
library(climateinsurance)
# List od input data
year <- c(1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015)
yield <- c(0.69, 1.62, 1.04, 1.11, 1.1, 1.03, 1.22, 0.49, 1.65, 1.01, 1.46, 1.1, 0.97, 0.82, 1.17, 1.38, 1.25, 0.75, 1.41, 0.66, 1.79, 1.1, 0.85, 1.21, 1.19)
index <- c(70, 137, 110, 114, 77, 90, 85, 42, 162, 127, 142, 114, 100, 76, 122, 112, 119, 74, 105, 45, 158, 107, 70, 122, 92)
# This package require to enter list of years of observation, crop yield
# information, index for insurance design, price of crop yield and stricke level
# for quantile regression. By default crop_price = 1 and strike_quantile = 0.3
climateinsurance(year, yield, index)
year index yield index_shortage yield_shortage revenue_losses payout premium non_insured_revenue insured_revenue he_insurance he_noinsurance
1 1991 0.69 70 0.324 16 16 27.2389 6.2952 70 90.9437 142.4749 1081.0944
2 1992 1.62 137 0.000 0 0 0.0000 6.2952 137 130.7048 0.0000 0.0000
3 1993 1.04 110 0.000 0 0 0.0000 6.2952 110 103.7048 0.0000 0.0000
4 1994 1.11 114 0.000 0 0 0.0000 6.2952 114 107.7048 0.0000 0.0000
5 1995 1.10 77 0.000 9 9 0.0000 6.2952 77 70.7048 1035.2449 669.7744
6 1996 1.03 90 0.000 0 0 0.0000 6.2952 90 83.7048 367.6891 165.8944
7 1997 1.22 85 0.000 1 1 0.0000 6.2952 85 78.7048 584.4413 319.6944
8 1998 0.49 42 0.524 44 44 44.0531 6.2952 42 79.7579 534.6326 3706.3744
9 1999 1.65 162 0.000 0 0 0.0000 6.2952 162 155.7048 0.0000 0.0000
10 2000 1.01 127 0.004 0 0 0.3363 6.2952 127 121.0411 0.0000 0.0000
11 2001 1.46 142 0.000 0 0 0.0000 6.2952 142 135.7048 0.0000 0.0000
12 2002 1.10 114 0.000 0 0 0.0000 6.2952 114 107.7048 0.0000 0.0000
13 2003 0.97 100 0.044 0 0 3.6991 6.2952 100 97.4039 29.9877 8.2944
14 2004 0.82 76 0.194 10 10 16.3097 6.2952 76 86.0145 284.4446 722.5344
15 2005 1.17 122 0.000 0 0 0.0000 6.2952 122 115.7048 0.0000 0.0000
16 2006 1.38 112 0.000 0 0 0.0000 6.2952 112 105.7048 0.0000 0.0000
17 2007 1.25 119 0.000 0 0 0.0000 6.2952 119 112.7048 0.0000 0.0000
18 2008 0.75 74 0.264 12 12 22.1947 6.2952 74 89.8995 168.4942 834.0544
19 2009 1.41 105 0.000 0 0 0.0000 6.2952 105 98.7048 17.4325 0.0000
20 2010 0.66 45 0.354 41 41 29.7611 6.2952 45 68.4658 1184.3344 3350.0944
21 2011 1.79 158 0.000 0 0 0.0000 6.2952 158 151.7048 0.0000 0.0000
22 2012 1.10 107 0.000 0 0 0.0000 6.2952 107 100.7048 4.7316 0.0000
23 2013 0.85 70 0.164 16 16 13.7876 6.2952 70 77.4924 644.5308 1081.0944
24 2014 1.21 122 0.000 0 0 0.0000 6.2952 122 115.7048 0.0000 0.0000
25 2015 1.19 92 0.000 0 0 0.0000 6.2952 92 85.7048 294.9882 118.3744
Call:
quantreg::rq(formula = calc_table$yield ~ calc_table$index, tau = strike_quantile)
Coefficients:
(Intercept) calc_table$index
0.8053097 84.0707965
Degrees of freedom: 25 total; 23 residual
$quantile_for_strike
[1] 0.3
$strike_level
[1] 1.014
$number_of_years
[1] 25
$number_of_losses
[1] 8
$number_of_payouts
[1] 8
$total_losses
[1] 149
$total_payouts
[1] 157.3805
$insurance_premium
[1] 6.29522
$ts
[1] 0.6
$pod
[1] 0.75
$far
[1] 0.25
$correlation
[1] 0.8787666
$rHE
[1] 0.5609766
Moreover, based on this packege web-app has been developed under the link https://klimalez.org/climate-insurance
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent