MODELING THE IMPACT OF CLIMATE VARIABILITY ON LOCAL FOOD PRICES - Food Security, Food Prices and Climate Variability - page 147

Kenya Nairobi Maize 13.7 13.9 - 0.04 - 13.5 0.04 -0.33 0.83 -0.04

Kenya Garissa Maize 21.2 22.3 19.4 0.04 -1.01 - - - - -

Malawi Bangula Maize 18.6 18.0 - 0.10 - 17.3 0.10 -0.80 0.77 -0.08

Mozambique Chokwe Maize

(white)

22.9 22.2 - - - - - - - -

Mozambique Angonia Maize

(white)

22.5 19.5 - - - - - - - -

Mozambique Gorongosa Maize

(white)

25.7 25.6 - - - - - - - -

Mozambique Manica Maize

(white)

17.6 13.8 - - - - - - - -

Mozambique Maputo Maize

(white)

12.9 13.3 - - - - - - - -

Mozambique Maxixe Maize

(white)

15.3 15.8 - 0.08 - 14.3 0.08 -1.09 0.64 0.12

Mozambique Nampula Maize

(white)

17.3 15.7 - - - - - - - -

Mozambique Nampula Maize

(white)

17.4 14.5 - - - - - - - -

Mozambique Ribaue Maize

(white)

23.8 19.3 - - - - - - - -

Rwanda Kigali Maize 21.0 20.1 - - - - - - - -

Tanzania Arusha Maize 16.6 11.9 11.7 0.02 -4.09 - - - - -

Tanzania Dar es

Salaam

Maize 11.8 11.0 - - - - - - - -

Tanzania Mbeya Maize 11.9 - 9.8 0.09 -4.36 - - - - -

Tanzania Songea Maize 24.3 18.7 16.5 0.01 -1.34 16.4 0.01 -1.57 0.85 -0.06

Notes
1 . Root mean square percent error between predicted 2009-12 prices and actual prices.
2 . P value of NDVI in model.
3 . Coeficient for NDVI in model.
4 . P value of world prices in model.
5 . Coeficient for world prices in model.
6 . Results for each model are reported when input elements have a signiicant p value, even if RMSPE is higher than the naïve model target.
7 . For millet, we used maize international prices for the “world price” in the model.