Forecasting And Technical Analysis Expert | Accredited

Explore forecasting techniques for investors and businesses to anticipate demand for goods and services while grasping inflation, deflation, and bubbles in the economic domain.

Explore how tulip bulbs became an infamous 17th-century financial bubble in Holland, illustrating what a bubble is and why recognizing it matters for forecasting.

Trace how the first recorded bubble forms across markets—from tulip bulbs in Holland to real estate—driven by demand shocks, flipping, and speculation that push prices higher.

Explore bubble formation driven by speculative demand and panic selling as major players exit. Understand the anatomy of a bubble and its bursting in real estate, tulips, and bitcoin.

The lecture demonstrates the bubble anatomy across bitcoin and the 2001 dot-com bubble, showing uptrend, rapid acceleration, burst, downward correction, pessimism, stabilization, and return to an old trend.

Learn to distinguish price spikes that reflect consumption from those driven by speculative demand, and anticipate whether a bubble will burst or prices may persist or rise.

Analyze whether Rome's 80 percent wheat price spike is a bubble by testing alternatives such as population, inflation, supply, and substitutes like corn or pasta.

Explore how the CPI measures inflation by comparing a goods basket to a base year, creating an index like 100 or 110. Recognize the PPI and basket criticisms.

Explore deflation, a general decline in prices, and contrast it with inflation. Examine how deflation can affect unemployment, income, consumption, and debt, and why it may harm long-term prospects.

Explore demand patterns for forecasting, including horizontal, seasonal, cyclical, random, and trend patterns, with real examples like Nintendo’s playing cards, ice cream, and EVs.

Explore how the SR-71 Blackbird and correlation compare to regression in forecasting, using high-speed, high-altitude data to build forecast models in a Cold War context.

Forecast Mars terraforming outcomes, including atmosphere, water, and soil for life. Assess how Mars' gravity and radiation affect human physiology, using company data to predict volunteers.

Explore how the correlation coefficient r measures the strength and direction of a linear relationship, with the Pearson product moment method, range -1 to 1, and software tools.

Explore the Mars data file from 2030–2043, detailing passengers (thousands), net monthly income, inflation rate not relevant, ticket price, living and personal space, health check, training duration, and Mars income.

Visualize the data with a scatter plot to assess the relationship between passenger numbers and monthly income. Add axis titles and check for outliers before using the correlation coefficient.

Analyze a dataset in Excel to compute single correlation coefficients and a correlation matrix, using the Correl function to assess relationships like passengers versus average net monthly income.

Learn to generate a correlation matrix in Excel with the Data Analysis Toolpak, including first-row labels, and interpret coefficients like 0.3 and -0.6 to reveal variable relationships.

Inspecting the matrix shows how to scan data for correlations. Focus on average net monthly income, inflation rate, ticket price, living space on Mars, and personal space on board.

Learn how correlation captures linear relationships, its sensitivity to outliers, and how it does not imply causality, with examples and guidance for distinguishing empty correlation from causal links.

Respect outliers by evaluating when to remove or keep them, using scatter plots and correlation coefficient to explore temperature and visitor data across two scenarios.

Explore correlations between a mock stock price and variables like short-term interest rates, the S&P 500, unemployment rate, long-term interest rate, and crude oil price, and identify notable directions.

Build a correlation matrix with the data analysis Toolpak, include labels, and reveal strong positive correlations between unemployment rate and stock price and between crude oil price and ABC 123.

Explore how data delays affect relationships between unemployment and discount-store revenue, and learn to test time lags of one or more periods to uncover true correlations.

Explore how time lags affect the link between promotional expenses and sales by computing correlations for lag 0, 1, 2, and 3 to identify the best lag.

Evaluate how time lag affects the correlation between promotional expenses and sales, from zero to two periods, noting a correlation at lag two and that the data was cooked.

Explore Sir Isaac Newton’s Cambridge professorship, his quiet genius, and how he began collecting student data to forecast performance in his mathematics course, with the attached data file to download.

Explore a data file of 30 students by Mr. Newton, detailing hours of lectures and self-study, extracurricular activities, an estimated IQ, and grades 0–100.

Explore single OLS regression, quantifying the effect of IQ on grade with a linear function defined by a slope and a constant to forecast, using Excel for computation.

Examine a single OLS regression and its slope and intercept, and interpret r-squared as the share of variance explained, with thresholds and caveats.

Analyze the correlation matrix to identify grades predictors, noting strong links with IQ and self-study hours, while extracurricular activity shows a weak relationship, guiding a forecast model centered on IQ.

Plot a scatter plot of IQ against grade in Excel, perform an OLS regression via a trendline, and note a slope of 1.83 and intercept -167 explaining 84% of variance.

Explore residuals in regression by distinguishing homoskedastic from heteroskedastic distributions and why checking residual spread matters for model accuracy.

When heteroskedasticity appears, don't discard the regression; use the model for well-explained regions while noting larger residuals at higher x, and apply judgment since no universal thresholds exist.

Demonstrate how earnings per share significantly predict stock price, with an R-squared of 0.96, while the intercept may be insignificant, yet the relationship remains informative.

Develop a forecast model using a single OLS regression to predict stock price from crude oil price, evaluate with R squared and heteroskedasticity, and forecast at oil price of $120.

Use the data analysis toolpak to run a single OLS regression of stock price on oil price, forecast at 120, and confirm significance with r-squared 0.98 and p-values below 0.05.

Explore how online schooling evolved from ascii art to faster loads and streaming, and apply nonlinear regression to market research data for building your online school.

Study nonlinear regression by examining how course duration affects panel scores, from 0.5 to 6.9 hours, with scores between 1 and 10 guiding model insights.

Explain why ols regression may underfit scatter data and introduce non linear regression options, like exponential and other curved trend lines, to improve forecasting.

Polynomial regression serves as a nonlinear alternative to linear regression, showing how course hours relate to satisfaction and achieving an R square of 0.92, with a 10-hour example.

Implement a logarithmic regression to explain satisfaction data, achieving an R^2 of 0.99 with y = 1.46 ln(x) + 3.11. Compute ln in Excel to predict ten-exercise satisfaction around 6.5.

Apply power regression in Excel to fit a power function and compare with linear trends; the method explains up to 97% of variance and predicts 6.3 for ten exercises.

Demonstrate exponential regression for course hours and satisfaction scores, compare r-squared with linear regression, and compute predictions using the exp function in Excel.

Examine polynomial regression, from order two to five, and how higher orders raise R square while not making sense outside the data domain, urging critical interpretation.

Model the link between festival temperature and visitor counts using nonlinear regression, comparing exponential, logarithmic, power, and polynomial options. Use a second-order polynomial to explain 92% of the R square.

Learn the basics of forecasting techniques by tracing Nintendo’s evolution from handmade hanafuda to a global video game leader, and how production relies on demand forecasts.

Explore the naive forecasting method, learn to calculate forecast errors and mean absolute deviation (MAD), and interpret overestimation and underestimation using a Nintendo demand example.

Learn to measure forecast accuracy with mean absolute percentage error (MAPE), interpreting percent deviations between actual demand and forecast, and aiming to minimize MAPE using various forecasting techniques.

Apply simple moving average by averaging the last two periods to forecast March and April, compare errors and MAPE with naive methods, and examine time factor choices.

Apply weighted moving averages with a smoothing factor and weights summing to one to emphasize recent data, and compare forecasts and MAPE on the Nintendo cards dataset.

practice a three-period simple moving average and a three-period weighted moving average with weights 0.6, 0.3, 0.1 to predict demand and compare them using MAPE to identify the best model.

Analyze simple moving averages and a three-period weighted moving average to forecast demand, compute errors and mape, and compare methods to identify the best forecasting approach.

Master the short selling principle: borrow an asset, sell at 100, buy back at 70 for a 30 profit if the price falls, while weighing risk and insurance.

Explore bullish and bearish engulfment candlestick patterns and how they signal shifts in market sentiment, indicating potential reversals in technical analysis.

Identify hammer candlesticks as reversal signals from bearish to bullish in candlestick charts, with long lower wicks and short bodies. Type A is usually a stronger signal than type B.