Optimal Pre-Harvest Grain Futures Marketing for Farmers
Farmers must meet liquidity demands for loan repayment, land rent, and other expenses at harvest. To meet these demands, they can either accept the cash price at harvest, which tends to be the lowest price period of the year, or they can sell futures contracts before harvest to hedge their risk and guarantee a certain level of revenue. Factors affecting this decision include price movements, global grain stocks, prices or related commodities, and expected yield. The underlying price factors (short-term mean price, stocks, volatility) are stochastic, and can be modeled using state variables with seasonality and mean reversion. We are currently implementing our model on price data for various commodities and are testing various objective functions for performance. We hope to publish this research early in 2017.
Effects of Tillage on Sorghum Yield and Soil Structure
Collaborators: Dr. David Clay, Justin Diabri
To test the effects of tillage on sorghum in Burkina Faso, we designed a randomized repeated block trial with four types of tillage including hand hoeing, animal-drawn plowing, tractor-drawn disking, and tractor-drawn plowing. We repeated each treatment three times in a randomized order. Each of the twelve resulting plots was tested for soil bulk density at the 0-10 cm and 10-20 cm depths before and after tillage. We planted traditional red sorghum after tillage, and the plots will be ready for harvest in late November. We plan to submit our results to the International Journal of Agronomy.
Stock Index Options Pricing Models
Collaborators: Dr. Zhiguang Wang, Dr. Jung-Han Kimn
This was my senior mathematics thesis which I conducted under the guidance of Dr. Jung-Han Kimn of the SDSU Department of Mathematics and Statistics and Dr. Zhiguang Wang of the SDSU Department of Economics. You can download the paper here.
Satellite-Derived Crop Insurance in the Developing World
Collaborators: Marcel Yanogo, Basepe Wepia
In 2012, I spent 4 months working as an extension agronomist in the country of Burkina Faso in West Africa. While there, I noticed that most farmers did not have access to credit, largely due to the farmers’ complete lack of risk management tools such as crop insurance. In interviewing farmers, bankers and insurers, I put together a picture of the challenges facing the industry, and of possible solutions. I used local government agricultural and meteorological data to identify the extent of the problem and the possible benefits from addressing it. You can download a summary of my findings here and appendices here (in French), or reach out to me via email for more information.
Variable Fertilizer and Seeding Rates
Collaborator: Dr. Gregg Carlson
Starting in 2010, I began conducting variable rate seeding and fertilizer experiments with Dr. Gregg Carlson of the South Dakota State University Department of Plant Science. Agricultural fields have continuously variable soil types, nutrients, organic matter, and susceptibility to floods and droughts. These variable conditions means that optimal seeding and fertilizer rates vary across fields as well.
To solve this problem, I designed two repeated-strip trials of various planting populations (22,000, 27,000, 32,000, 37,000 and 42,000 seeds per acre) and nitrogen fertilizer rates (70 lbs, 100 lbs, 130 lbs, 160 lbs and 190 lbs per acre) for corn fields on my home farm in Lake Wilson, MN. I wrote programs to ingest historical yield monitor data and divide each field into regions of similar yield potential through the use of cubic splines. Within each yield potential region, we determined the yield response to the treatment through various linear regression models.
We found the best model to be a second order polynomial for both seeding rate and nitrogen. Using our yield response models and market prices for corn, seed and nitrogen fertilizer, we determined the overall optimal seeding and nitrogen fertilizer rates within each yield potential region. Contact me if you would like to learn more.