Polynomial Regression and Yield Response Calculator: Quadratic, Plateau Models and Economic Optimum Dose
Enter the dose (for example kg N/ha) and the yield of each plot. The calculator fits linear, quadratic, cubic, linear-plateau, quadratic-plateau and Mitscherlich (exponential rise to a maximum) models, compares them by AICc, R² and residual error, and calculates the agronomic optimum (maximum yield) and the economic optimum dose from your fertiliser and crop prices, with 95% confidence intervals.
- PriceFree, no login or sign-up needed
- Your dataProcessed in your browser, never uploaded
- Works onPhone, tablet or computer, any modern browser
- Made byDr. Muhammad Aammar Tufail, Codanics
- Published17 November 2026
- Last updated16 November 2026
📚 How to cite this tool
This tool is free. If it helped your thesis, paper, report or assignment, please cite it. It keeps the tools free and helps other students find them.
Tufail, M. A. (2026). Polynomial regression and yield response calculator with economic optimum dose [Online tool]. Codanics. Retrieved 16 November 2026, from https://codanics.com/tools/polynomial-regression-calculator/
In-text citation: (Tufail, 2026) · Narrative: Tufail (2026)
The access date is today’s date, filled in automatically. Author: Muhammad Aammar Tufail. Published 17 November 2026.
How to use the Polynomial and Yield Response Calculator
- Enter one plot per line: the dose (for example kg N/ha), then the yield. Keep replicates on separate lines. A simulated wheat nitrogen trial (7 rates × 4 replicates) is loaded as an example.
- Type the price of one dose unit (for example $ per kg N) and the price of one yield unit (for example $ per tonne of grain) in the same currency.
- Click Fit models. The model with the lowest AICc is shown; switch to another model in the list if it makes more agronomic sense.
- Read the agronomic optimum (maximum yield), the economic optimum dose and their 95% CIs, and compare the models in the table and the chart.
- Copy the tables or the results sentence, download the CSV and save the chart as PNG or SVG.
Why the response model matters
Crop yield usually rises with fertiliser, irrigation or seed rate and then levels off. How you describe that curve decides the recommended dose. A quadratic model always curves down after its maximum and often recommends more fertiliser than a plateau model fitted to the same data, while a Mitscherlich curve approaches a maximum smoothly without reaching it. Agronomists therefore fit several models and compare them before recommending a rate (Cerrato & Blackmer, 1990).
The models
| Model | Equation | Optimum |
|---|---|---|
| Linear | y = a + b·x | None: yield keeps rising over the doses tested |
| Quadratic | y = a + b·x + c·x² (c < 0) | Maximum at −b/(2c); economic optimum (r − b)/(2c) |
| Cubic | y = a + b·x + c·x² + d·x³ | Where b + 2c·x + 3d·x² = 0 (or = r) |
| Linear-plateau | y = a + b·min(x, x₀) | Join point x₀ (critical dose), if the slope b exceeds the price ratio |
| Quadratic-plateau | quadratic up to x₀ = −b/(2c), then flat | x₀; economic optimum (r − b)/(2c), never above x₀ |
| Mitscherlich | y = a − b·e^(−c·x) | No true maximum: dose for 95% of the asymptote = ln(20)/c; economic optimum ln(b·c/r)/c |
Economic optimum dose
The economic optimum nitrogen rate (EONR) is the dose at which one more unit of fertiliser increases yield by just enough to pay for it. Mathematically, the slope of the response curve equals the price ratio r = price of one dose unit ÷ price of one yield unit. With N at $1.10/kg and wheat at $250/t, r = 0.0044 t of grain per kg of N. A higher fertiliser price or lower grain price lowers the economic optimum.
How the fits are calculated
The polynomial coefficients match R’s lm(y ~ x + I(x^2)), and the plateau and Mitscherlich parameters and standard errors match nls() / minpack.lm::nlsLM() to at least six significant digits on the example data.
Tips for fertiliser trials
- Use at least five or six rates, including zero and one clearly above the expected optimum, so the plateau can be estimated.
- Analyse each site-year separately first; pool only when the responses are similar.
- Test the treatment effect with the field-trial ANOVA calculator (RCBD) and use this tool to describe the dose response.
- Report the model, its parameters with SEs, the optimum doses with CIs and the price ratio used.
Frequently asked questions
What is polynomial regression?
Polynomial regression fits a curve by adding powers of the predictor (x², x³) to a linear model. It is still estimated by least squares, so the coefficients have standard errors, t-tests and confidence intervals. A quadratic (second-order) polynomial is the classic model for a response that rises and then falls.
Which model should I choose?
Start with the lowest AICc, but also use agronomic sense. If two models differ by less than about 2 AICc units, both fit equally well. Plateau models usually give lower and more conservative optimum doses than the quadratic model; many studies prefer the quadratic-plateau or linear-plateau model for nitrogen.
How is the economic optimum calculated?
It is the dose at which the slope of the fitted curve equals the price ratio (price of one dose unit ÷ price of one yield unit). For a quadratic y = a + bx + cx², it is (r − b)/(2c). Enter both prices in the same currency and make sure the yield price matches the yield unit (per tonne if yield is in t/ha).
Why does the linear-plateau model give a sharp corner?
Because it assumes yield rises at a constant rate until the critical dose and then stays flat. Real responses curve more gradually, but the model is simple, often fits well and gives a clear critical dose.
Why can't the plateau models be fitted to my data?
If the yield keeps rising up to the highest dose, there is no plateau within your data and the join point cannot be estimated. Use the linear, quadratic or Mitscherlich model, or include higher doses in your next trial.
Can I use it for other dose responses?
Yes: irrigation amount, seed rate, compost, salinity or temperature responses all work. For sigmoidal dose responses with an IC50 or EC50, use the dose–response (IC50) calculator instead.
Does it match R or SAS?
Yes. The polynomial models match lm() in R and PROC REG in SAS; the plateau and Mitscherlich models match nls() in R and PROC NLIN in SAS. Small differences can appear in the last digits of nonlinear fits.
Do I need to log in or create an account to use the Polynomial and Yield Response Calculator?
No. There is no login, no sign-up and no email required. Open the page and start using it straight away, for free, on any device.
How do I cite this tool in my thesis or paper?
Use the citation box on this page. In APA 7th edition: Tufail, M. A. (2026). Polynomial regression and yield response calculator with economic optimum dose [Online tool]. Codanics. Retrieved [date you used it], from https://codanics.com/tools/polynomial-regression-calculator/. BibTeX, RIS (EndNote, Mendeley, Zotero), MLA, Harvard, Chicago and Vancouver formats are also available.
Free guides and courses from Codanics
More free tools
See all free Codanics tools → · Codanics on YouTube · WhatsApp channel

