1 / 28

Risks and return periods

Risks and return periods. Module I3 Sessions 8 and 9. Learning objectives. From this session you should be able to: Generalise the 5-number summary to give any percentile, or risk level Explain risks in a variety of ways, to suit different users.

dreama
Download Presentation

Risks and return periods

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Risks and return periods Module I3 Sessions 8 and 9

  2. Learning objectives • From this session you should be able to: • Generalise the 5-number summary • to give any percentile, or risk level • Explain risks in a variety of ways, • to suit different users. • Be able to interpret a cumulative frequency curve to specify • values for a given risk, • and risks for a given value.

  3. Climate risks and other risks • People have to take risks • If they knew the size of the risk • e.g. 1 year in 10 • or 10% chance • They would have the information • To plan their action • Without this information • they have to guess • often conservatively • sometimes rashly • Can we help? • By interpreting the variability • As statements of risk • That people can use

  4. Contents • Activity 1: This presentation • Activity 2: Peter Cooper interview • climate risks • Activity 3: Demonstration of risks in CAST • Activity 4: Practical 1 • the results from the interview • learning about risks and return periods – CAST • Activity 5: Practical 2 • Calculating risks in Excel • To estimate the chance of solar cooking • Using sunshine data • Activity 6: Review • Summarising data well

  5. From DFID key sheet 6 2004, www.dfid.gov.uk

  6. Peter Cooper ICRISAT Linking current climatic variability (using the historical data) to climate change

  7. Activity 2: Interview with Peter Cooper • Particularly the points about data • And risks for farmers • Discussed on the next slides Watch the interview or use the transcript

  8. An example of “accelerated learning, using historical climate data (Masvingo, Zimbabwe) • Nitrogen recommended for maize (52kg/ha) • but not adopted • Why not? • Too expensive and thought to be too risky. • We asked “how much could farmers afford”? • The answer was about 17kg N /ha • ‘Risk and returns’ analyses was done • by a crop simulation program (APSIM) • with 47 years of daily historical climate data • To compare the risks • with no fertilizer, 17kg and 52kg

  9. An example of “accelerated learning” - results Fertilizer “failed” in a few years Sometimes it increased yields a little “Usually” it increased yields a lot

  10. Probability or ‘Rates of Return’(how many years out of ten?) rate of return on 17 kg/ha rate of return on 52 kg/ha Result: Good probability of higher rate of return with lower inputs • Impact: Extension Services, Fertilizer Traders and ICRISAT, • recently successfully evaluated nitrogen “micro-dosing” • with 200,000 farmers in Zimbabwe – still expanding.

  11. Activity 3: Cast and risks • Some risks have already been seen • For example with boxplots • See next slide • CAST has a new chapter on risks and reurn periods • Which we review here • And then investigate • in Practical 1 • How should you phrase risks • So they are easily understood

  12. Starting points – some risks already Boxplots and risks – sessions 2/3

  13. CAST and risks

  14. Expressing risks in appropriate ways percentage probability rate

  15. Exercises too!

  16. Activity 4: Practical 1 • Results from the Peter Cooper interview • Then work on CAST • Working in pairs • To practice explanations to your partner • If you can explain a topic • and explain your reasoning • Then you probably understand it

  17. Solar cooking Case study:Sunshine data • Module B1 Session 8 describes the problem • Here we examine the risk of not having enough sun • Data: • The raw data has be made into variables for analysis • But they are still available as in the Zambia rainfall data • Objectives: • Find proportion of days when cooking is possible • Find whether sunshine early morning is related to this proportion • So can you reduce the risk, by knowing the state of early part of the day?

  18. Adopting a solar cooker • The context • Adopting a solar cooker • depends on many things, some statistical, others not • One statistical aspect • What proportion of days can it be used? • We therefore analyse the data to find out • What is the risk? • Then, use the ideas from Session 7 • Can we reduce the unexplained variability? • By using a related measurement, - early morning sunshine • We can not reduce the risk • But we can reduce the (last minute) risk • And hence help people to be able to plan better

  19. Activity 5: Practical 2 – Excel for risks Getting the summary Plotting the summary % of days with < 3hrs

  20. Learning objectives From these sessions you should be able to: • Generalise the 5-number summary • to give any percentile, or risk level • Explain risks in a variety of ways, • to suit different users. • Be able to interpret a cumulative frequency curve to specify • values for a given risk, • and risks for a given value.

  21. Review • Now you have the tools and skills to process • Factors (categorical or qualitative data) • Using frequencies, proportions and percentages • Variates (quantitative data) • Using means and medians • And quartiles, extremes, standard deviations • And proportions (risks), percentiles (return periods) • You also know to use other measurements • to reduce the unexplained variation

  22. Variability shown graphically (Sessions 2/3) Variability can partly be explained by the variety

  23. Variation shown numerically (Sessions 4/5) You can interpret measures of variation including s.d. So are able to picture the data if given a summary

  24. Limitations of each summary statistic, e.g.

  25. Sessions 6/7: Reducing unexplained variation If it could be done as well as this, then seasonal forecasting would be in good shape! Variation after forecast Overall variation The Analysis of variance was introduced Showing both the variance and s.d. are used

  26. Applying the information • Swaziland crop cutting survey • Further analysis to be done • To examine the relationships • between yield of maize and various inputs • like fertilizer and variety • What might cause variation? • Perhaps early planting is an important variable? • Currently it has not been measured • In the future • The discussions on explaining variation • Are leading to its measurement from now on!

  27. In Sessions 8 and 9 here we looked at risks The cumulative frequency curve is to be interpreted Risks can be stated in different ways

  28. Next we see how all these summaries can be displayed in tables

More Related