Help me build and app in streamlit that will guide users trhough a workflow of selecting the best statistical test to apply to their use case. It should start by asking them to indicate the problem. i.e. First ask: Hypothesis Testing: For when you want to evaluate a claim or theory about a population parameter, such as the mean or proportion. Example: A company claims that their new battery lasts, on average, more than 10 hours. You can use a hypothesis test (e.g., a one-sample t-test) to determine if the sample data supports or refutes this claim. Comparing Groups: For when you want to test if there is a significant difference between two or more groups or treatments. Example: A researcher wants to know if a new teaching method leads to better test scores compared to the traditional method. They can use a two-sample t-test or ANOVA to compare the mean scores between the groups. Relationship Analysis: For when you want to understand the strength and nature of the relationship between two or more variables. Example: A marketer wants to know if there is a relationship between advertising expenditure and sales revenue. They can use correlation and regression analysis to quantify and model this relationship. Sample Size Determination: For when you need to calculate the required sample size for a study or survey to achieve a desired level of precision or statistical power. Example: A pollster wants to estimate the proportion of voters favoring a particular candidate with a margin of error of 3% and 95% confidence. They can use sample size calculations to determine the minimum number of voters they need to survey. Quality Control: For monitoring and ensuring that a manufacturing or production process is operating within acceptable limits. Example: A factory uses control charts to monitor the weight of cereal boxes being produced. If the weights fall outside the control limits, it indicates a problem with the process that needs to be addressed. Forecasting and Time Series Analysis: For analyzing and predicting future values or patterns in data that is collected over time. Example: A retailer wants to forecast next year's sales based on past sales data. They can use time series models like ARIMA or exponential smoothing to identify trends and seasonality patterns in the data and make predictions. Experimental Design: For planning and designing experiments or studies in a way that minimizes bias and allows for valid conclusions to be drawn. Example: An agricultural researcher wants to study the effect of different fertilizers on crop yield. They can use principles of experimental design (e.g., randomization, blocking) to set up the experiment and ensure that any observed differences can be attributed to the fertilizers. Survey Analysis: For analyzing data collected from surveys or samples, accounting for potential biases and making inferences about the larger population. Example: A market research firm conducts a survey to estimate the proportion of consumers who prefer a particular product. They can use statistical methods to analyze the survey data, calculate confidence intervals, and make inferences about the overall consumer population. Risk Assessment and Reliability Analysis: For evaluating the probability of failures, losses, or other risks, and assessing the reliability of systems or products. Example: An engineer wants to estimate the likelihood of a structural failure in a bridge design over its lifetime. They can use probability distributions and reliability analysis techniques to quantify the risk and inform design decisions. Then based on the slection, ask things like: Hypothesis Testing: What are you trying to prove or disprove? Are you testing for a one-sided or two-sided effect? How picky do you want to be about calling something significant? Does your data look like it came from a normal bell-curve distribution? Comparing Groups: How many different groups are you comparing? Are the groups completely separate, or are they related somehow (like before and after)? Does your data look normal and have similar spreads across groups? If there is a difference, which specific groups are different from each other? Relationship Analysis: Is the relationship between your variables a straight line or more complicated? Are there any outliers or extreme values that might be skewing things? Does your data meet the typical assumptions for this type of analysis? Are there other factors that might be influencing the relationship you're looking at? Sample Size Determination: How precise or how big of an effect do you want to be able to detect? How much variability or spread do you expect in your data? How confident do you want to be in your results? Are you looking at one group, two groups, or something more complex? Quality Control: What limits do you want to use to flag a potential problem? How often should you check to see if the process is running smoothly? What should you do if the process seems to be going off track? Are there any special circumstances or events that might be causing issues? Forecasting and Time Series Analysis: Does your data have a consistent pattern over time, or does it drift or shift? Are there any trends, seasonal cycles, or repeating patterns? How far into the future do you need to predict? How much error or uncertainty in the forecast is acceptable? Experimental Design: What factors or variables are you testing? Are there any other things that might be affecting your results that you need to account for? How will you randomly assign treatments or conditions? How many times do you need to repeat each condition or treatment? Survey Analysis: How did you select the people you surveyed? Are there any reasons why some people might not have responded? Do you need to adjust or weight the results based on demographics? How precise or accurate do your survey estimates need to be? Risk Assessment and Reliability Analysis: What kind of failure or bad event are you trying to avoid? How much risk or chance of failure is acceptable? Are there any backups or safety nets built into the system? How will you monitor and maintain the system over time? and so on until it guides the user to tell/determine the type of data they need, and lastly provide it so the app can make the testings accordingly
