RiskQ 4.2

Software for Probability and Statistics

Available packages for statistics typically limit your flexibility. Many types of analyses are based on the "normality assumption" of the data, yielding results that are comprehensive and easy to manage, but do not reflect the real characteristics of the data. RiskQ puts you in control. Using empirical data, you can build and manipulate population distributions so you can conduct customized statistical tests. This is critical to performing more realistic assessments of field and experimental data that do not meet simple distributional assumptions. For example, imagine the distribution of flood data is not normal but has a fat tail. This tail strongly governs the probability of extreme floods, and it should not be ignored. RiskQ allows such distributions to be characterized, worked with, and used for statistical testing.

These are some examples of RiskQ's utility:
  • Toxicology
    • Test new drugs in lab rats for safety
    • Evaluate LD50s
    • Test environmental pollutants

  • Risk Analysis
    • Given historical data, evaluate the probability of extreme flood events
    • Perform failure analysis on systems or devices

  • Epidemiology
    • Given empirical distributions, evaluate the probability that a cancer cluster is other than chance
    • Compare mortality in two human populations on different diets or exposed to different levels of pollution

  • Actuarial Analysis
    • Evaluate the probability of extreme events and their costs for insurance determinations
    • Evaluate lifespan data for different populations for life insurance premiums

  • Clinical Trials
    • Test clinical populations for drug performance

  • Experimental Data Analysis
    • Tailor ANOVA and MANOVA to your particular distributional assumptions
    • Do non-standard statistics
    • Test treatment effects

RiskQ supports numerous continuous and discontinuous probability distributions, including user-supplied empirical ones, which may be operated on by RiskQ functions. Additional RiskQ functions facilitate the characterization of uncertainty and discrete probability calculus. RiskQ also contains a variety of parametric and nonparametric statistical functions.

Developed and supported by Loehle Enterprises.