Advanced Real Estate Big Data Analytics Flashcards

Comprehensive vocabulary flashcards for the Advanced Real Estate Big Data Analytics exam summary, covering forecasting methods, AI models, risk management, and portfolio construction.

Predictable

An expected outcome that can be estimated because the process or the set of outcomes is known, such as office market returns using historical evidence.

Forecastable

An outcome that is close enough to the actual outcome to be useful, measured with forecast errors such as $MSE$.

Unpredictable

A process that is unknown or defined by outcomes that are undefined, such as black swans, regime changes, or specific single transaction prices.

Judgmental Forecast

Subjective analysis of subjective inputs from executives, experts, or surveys, used when data is limited or structural change is high.

Time Series

A time-ordered sequence of observations taken at regular intervals, relying on historical observations of the target variable.

Associative Model

A forecasting approach that uses observable causal drivers (e.g., interest rates or GDP) to predict a target outcome rather than relying only on the target's past.

Trend

The long-term movement in data, such as the long-run increase in house prices.

Seasonality

Short-term regular variation in a time series, such as hotel occupancy by quarter or transaction spikes in Q2.

Cycle

Wavelike long-term variation in real estate markets, often occurring around peaks like 1989, 2007, and 2022.

Irregular Component

Unusual circumstances in data caused by specific shocks, such as the 2008 financial crisis or the COVID-19 pandemic.

Random Component

Chance noise or measurement variation consisting of period-to-period wiggles that are not predictable.

Moving Average

A technique for reducing random short-term noise by taking the mean of several recent observations.

Exponential Smoothing

A forecast update rule where F_t = F_{t-1} + \text{\alpha} \times [A_{t-1} - F_{t-1}]; low $\alpha$ means slow adjustment and high $\alpha$ means high responsiveness.

Appraisal Smoothing

A bias in private real estate where valuations anchored to historical prices make measured volatility look lower than true market volatility, often expressed as V_t = \text{\alpha} \times \text{market price} + (1 - \text{\alpha}) \times V_{t-1}.

Nowcasting

The prediction of the present or very near future using higher-frequency proxies (like Google Trends or nightlights) when official data is delayed.

ARIMA

A univariate time-series model that splits data into cycles and trends using only the target variable's own past by design.

Random Forest

A non-linear multivariate machine learning model consisting of many trees with randomization, used for messy drivers and robustness.

Gradient Boosting

A machine learning technique using sequential trees where each corrects previous errors; provides maximum tabular accuracy but is prone to overfitting.

Risk

A measure of the probability and consequence of uncertain future events, mathematically expressed as $\text{Risk} = \text{Probability} \times \text{Consequence}$.

Uncertainty

A state where the randomness of outcomes cannot be expressed in specific probabilities because the process is unknown.

Systematic Risk

Market-wide risk that cannot be diversified away, such as interest rates, inflation, and macro-recessions.

Idiosyncratic Risk

Asset-specific risk related to a particular tenant, building condition, or local micro-market.

Discount Rate

The rate used for a future cash flow, calculated as $d_t = r_{f,t} + r_{p,t}$, where $r_f$ is the risk-free rate and $r_p$ is the risk premium.

Due Diligence

The investigation performed before acquiring a property to discover information needed to assess if investment risk is suitable.

Mean

The expected value or central outcome representing the most useful single summary of a distribution.

Standard Deviation

The dispersion of outcomes around the mean; higher values indicate higher volatility and risk.

Sensitivity Analysis

A 'what-if' analysis that changes one input variable at a time to observe its effect on a target variable like IRR or NPV.

Scenario Analysis

A risk tool that changes a coherent set of assumptions together to capture how risks are correlated, such as in a 'worst case' recession.

Monte Carlo Simulation

A technique that estimates outcomes by drawing random input values from probability distributions and running the model thousands of times to produce a distribution of results.

Exit Yield

The capitalization rate used to estimate the terminal value of a property at the end of a holding period.

Modern Portfolio Theory (MPT)

A framework for combining assets to maximize return for a given level of risk or minimize risk for a given level of return through diversification.

Efficient Frontier

A plot representing portfolios that provide the best possible return for a specific level of risk.

Sharpe Ratio

A measure of excess return per unit of risk, calculated as $\frac{\text{portfolio return} - \text{risk-free rate}}{\text{portfolio risk}}$.

Indivisibility

A characteristic of real estate assets where properties cannot be divided perfectly into small fractional shares, affecting portfolio optimization.

Income Return ($q_i\_return$)

The quarterly income return calculated as $\frac{\text{mean rent} \times 3}{\text{price}}$.

Capital Return ($q_c\_return$)

The return from price appreciation, calculated as $\frac{\text{price}_t - \text{price}_{t-1}}{\text{price}_{t-1}}$.

Mean Square Error (MSE)

A forecast error metric calculated as the average of (Y_i - \text{\hat{Y}}_i)^2.