004-Forecasting Flashcards
This is the process of developing knowledgeable presumption about certain business measures
Business Forecasting
Elements of a good forecasting
- Simple to understand and use
- Flexible and adjustable
- Timely
- A line with strategic goal
- Accurate and reliable
A tool used by businesses to make well informed decisions in a variety of domains, including financial planning, operations, inventory management, and customer service.
By doing this businesses can improve productivity profitability and customer happiness to predict the future conditions and make appropriate preparations
Forecasting
A type of forecasting.
It predicts the demand for a product or service over a certain period
Demand Forecasting
A type of forecasting.
Predicts future sales volumes based on historical data, market trends, and promotional activities
Sales Forecasting
A type of forecasting
Involves projecting financial outcomes based on expected sales and operational cost
Financial Forecasting
A type of forecasting
Estimates the availability and reliability of supply sources, considering supplier capabilities, lead times, and other factors
Supply Forecasting
Advantages of Forecast
- Improve decision making
- Increase the efficiency of the use of resources
- Identify opportunities and weaknesses.
- Maintains customer satisfaction
Disadvantages of Forecasting
1.It can be costly
2. Limitations of forecasting models 3.rapid market changes
4. dependence on historical data
Six Steps in the Forecasting Process
- Determine the purpose of the forecast
- Establish a time horizon
- Select the forecasting technique
- Obtain, clean, and analyze appropriate data
- Make the forecast
- Monitor the forecast
A forecasting approach.
It relies on subjective judgment and non-numerical data to predict future events incorporating factors like emotions, experiences, behavioral patterns, an experts opinions.
Valuable for forecasting trends like new product adaptations or shift and consumer behavior prehistorical data may fall short
Qualitative Forecasting
A forecasting approach
Uses historical data or causal relationship between variables to make forecast it relies on numerical data to predict future trends.
Common methods like regression analysis, time series analysis, and econometrics are used in field such as finance in marketing
Valid for the objectivity and ability to handle large data sets
Quantitative Forecasting
A forecasting approach
The lies on subjective inputs. Including opinions from customers, survey, sales staff, managers, executives and experts to make predictions.
Used when quantitative data is unavailable, incomplete or difficult to interpret
It combines insights from individuals who have a direct experience or specialized knowledge of the market industry
Judgemental Forecasting…
* Executive Opinions: top management
* Salesforce Opinions: sales people
* Consumer Surveys:* data collected from consumers
Delphi Method: a panel of experts provide prediction anonymously.
A type of forecast where it is predicting future values based on previously observe data points collected at a regular time intervals.
It assumes that patterns or trends from the past will continue in the future
Time Series Forecast.
Methods used in time series for casting:
* Moving Averages: smooth out short term fluctuation by averaging data over a fixed period.
* Exponential Smoothing: use when there is no trend or seasonality. Gives more weight to recent observations making it responsive to changes and moving averages
* Arima (auto regressive integrated moving average): (using past values) difference ( to make the data stationary), and moving averages (to smooth the noise)
A time series behavior
It represents the long term movement or direction in the data either upward downward or flat.
It can be linear or nonlinear depending on how the data behaves over time
Trend
A time series behavior.
It refers to the regular predictable patterns that repeat at a fixed interval within a year, weeks or day
Occur due to external factors like whether holiday or cultural events
Seasonality
A time series behavior
Do not have a fixed periodicity and are usually influenced by economic or business conditions such as economic regression or booms.
It does not follow a regular pattern like seasonality
it also last several years
Cycle
Refers to the different patterns or components that can be observed in time series data.
These behavior held in identifying trends and making forecast
Time Series Behaviors
Also known as “noise”are the unpredictable and erratic fluctuations in the data that do not follow a pattern
Can be caused by natural disasters strikes or sudden market destructions.
These are difficult to forecast and often random in nature
Irregular Variations
Refers to the unpredictable random fluctuations in time series data that “cannot be attributed to trends, cycles, or seasonality”.
Arises due to chance factors and typically have no pattern or structure.
Example: an unexpected spike in demand for a product
Random Variation
One of the simplest forecasting methods.
The forecast for the next period is based on the value of the previous period.
It assumes that the most recent observation is the best predictor for the future
Naive Forecast
Involves predicting future values based on historical data.
Commonly used when the data tends to fluctuate around the constant level or an average.
This technique concept is that the future value is often expected to be similar to the past values, with this assumption that any irregularity in the data are random and don’t significantly affect long term trends
Time Series Forecasting-Averaging
- Moving Average: calculate the forecast for future value by averaging a set of number of past observation
*The Weighted Moving Average: giving equal weight to all past observations, different weights are assigned to each observation. Usually giving more importance to most recent data
- Exponential Smoothing:a technique that applies exponentially decreasing weights to pass of distribution. This technique gives more weight to recent data, and decreases weight mostly as you go back in time.
Applies multiple forecasting techniques to the most recent historical data and then select the one with the highest accuracy to generate the forecast for the next period.
Focus Forecasting
Used when there’s no historical data for a new product or innovation.
It “predict” future product adoption and spread based on social and behavioral factors, such as adoption rates, social influences, external factors like media attention and word of mouth
Diffusion Models
A forecasting technique that are used to “project” future values by analyzing the relationships between multiple variables.
Assumes that fluctuation in one more independent variables (predictors) can influence and help predict the value of the dependent variables(target)
Associative Forecasting Techniques
An associative forecasting method.
This is a statistical methods that is used to “examine” the relationship between dependent variable and one or more independent variables.
It aims to understand how changes in the independent variable impact the dependent variable.
Simple linear regression: there’s only one independent variables. Models a straight line relationship between dependent and dependent variable.
Regression Analysis
An associative forecasting method
Used to “assess” the strength and direction of the linear relationship between two variables: a dependent variable (the one we aim to forecast) and the independent variable (a potential predictor)
Correlation Analysis
Refers to the continuous process of tracking and reviewing actual performance and comparison to predicted outcomes.
The goal is to spot any discrepancies early and understand the reason for those differences, enabling timely adjustments to business strategies.
Monitoring Forecast
Involves taking corrective actions to ensure that actual performance stays aligned with the forecasted goals.
Controlling Forecast
Refers to the process of continuously monitoring and adjusting forecast to ensure their accuracy
Forecast Accuracy and Control
Measuring Forecast Accuracy
- Mean Absolute Deviation: measures the average magnitude of errors between forecasted and actual values.
- Mean Absolute Percentage Error: expresses forecast error as percentage
- Tracking Signal: a diagnostic tool used to detect bias and forecasting by comparing cumulative forecast errors to their MAD
