What is the primary difference between the linear transformations of $y = Ab^x$ and $y = ax^n$?

In $y = Ab^x$, only the y-axis is logarithmic ($\ln y$ vs $x$), whereas in $y = ax^n$, both axes are logarithmic ($\ln y$ vs $\ln x$). This determines whether the relationship is exponential or a power law.

When comparing two exponential models $y = 5e^{0.2t}$ and $y = 5e^{-0.5t}$, which represents decay and why?

The model $y = 5e^{-0.5t}$ represents decay because the coefficient in the exponent is negative. This causes the value of $y$ to decrease toward zero as $t$ increases.

How does the interpretation of the gradient differ between a graph of $\ln y$ vs $x$ and a graph of $\ln y$ vs $\ln x$?

In a $\ln y$ vs $x$ graph, the gradient represents the natural log of the base ($\ln b$). In a $\ln y$ vs $\ln x$ graph, the gradient directly represents the power or exponent ($n$) of the original function.

What is a common error when finding the initial value $A$ from a linearized log graph?

Students often mistake the y-intercept $c$ for the initial value $A$. Because the graph plots $\ln y$, the intercept is actually $\ln A$, so the initial value must be calculated as $A = e^c$.

Why is it incorrect to assume an exponential growth model will be accurate for all future time values?

Exponential models assume unlimited resources and no constraints. In reality, factors like space, food, or competition create a 'carrying capacity' that eventually slows or stops growth, making the model invalid for large $t$.

What happens to the calculation of $k$ if you use $\log_{10}$ instead of $\ln$ when linearizing $y = Ae^{kt}$?

Using $\log_{10}$ will still produce a straight line, but the gradient will be $k \log_{10} e$ instead of just $k$. This adds an extra step of division to isolate the rate constant $k$.

Define the 'Rate Constant' $k$ in the context of the model $y = Ae^{kt}$.

The rate constant $k$ is the proportional growth rate. It represents the constant of proportionality between the current value $y$ and its instantaneous rate of change $\frac{dy}{dt}$.

What is the mathematical identity used to convert $a^x$ into a base-$e$ expression?

The identity is $a^x = e^{x \ln a}$. This is derived from the fact that $a = e^{\ln a}$, and then applying the power rule of indices.

In the linear form $\ln y = mx + c$, what do $m$ and $c$ represent for the model $y = Ab^x$?

The gradient $m$ represents $\ln b$ (the natural log of the growth base), and the intercept $c$ represents $\ln A$ (the natural log of the initial value).

Why does the graph of $y = Ae^{kt}$ never cross the x-axis?

The expression $e^{kt}$ is always positive for any real value of $k$ and $t$. As $t$ becomes very large and negative (for growth) or very large and positive (for decay), the value approaches but never reaches zero.

Library Podcasts

Courses

Referral & Rewards

Exponentials & Logarithms

Modelling with Functions

Summary

Mathematical modelling uses functions to represent real-world phenomena, particularly exponential growth and decay. By applying logarithmic transformations, non-linear relationships can be converted into linear forms, allowing for the estimation of parameters and the prediction of future trends.

1. Definition & Core Concepts

Graph showing exponential growth (red curve rising) and exponential decay (blue curve falling) starting from initial values on the y-axis.

2. Underlying Principles

3. Methods: Linearization using Logs

4. Key Distinctions

5. Exam Strategy & Tips

Modelling with Functions

Summary

1. Definition & Core Concepts

Exponential Modelling involves using functions of the form $y = Ae^{kt}$ or $y = Ab^x$ to describe quantities that change at a rate proportional to their current value. These models are essential in fields such as biology for population growth, finance for compound interest, and physics for radioactive decay.

The Initial Value is represented by the constant $A$ , which corresponds to the value of the dependent variable when the independent variable (usually time $t$ ) is zero. In a graph, this is the y-intercept where the curve begins.

The Rate Constant $k$ determines the speed of change; if $k > 0$ , the function represents exponential growth, whereas if $k < 0$ , it represents exponential decay. The magnitude of $k$ dictates how steeply the curve rises or falls over time.

Graph showing exponential growth (red curve rising) and exponential decay (blue curve falling) starting from initial values on the y-axis.

2. Underlying Principles

The principle of Rate of Change is central to these models, where the derivative $\frac{dy}{dt}$ is proportional to $y$ . For the growth model $y = Ae^{kt}$ , the rate of change is $\frac{dy}{dt} = Ake^{kt} = ky$ , meaning the faster the population grows, the more individuals there are to reproduce.

Exponential functions and Natural Logarithms are inverse operations, defined by the relationship $e^{\ln(x)} = x$ . This property allows us to solve for unknown exponents by 'taking logs' of both sides of an equation, effectively bringing the variable down from the exponent level to the base level.

Any exponential base $a^x$ can be rewritten in terms of the natural base $e$ using the identity $a^x = e^{x \ln a}$ . This standardization is useful because the calculus of $e^x$ is simpler than that of other bases, making it the preferred form for complex modelling.

3. Methods: Linearization using Logs

Linearization is the process of transforming a non-linear power or exponential equation into a linear form ( $Y = mX + c$ ) to make data analysis easier. By plotting logarithmic values instead of raw values, curved data points align into a straight line, allowing for the use of linear regression.

To transform $y = Ab^x$ , take the natural log of both sides to get $\ln y = \ln(Ab^x) = \ln A + x \ln b$ . In this linear form, the gradient is $\ln b$ and the vertical intercept is $\ln A$ , with the independent variable remaining as $x$ .

To transform a power model $y = ax^n$ , take logs to get $\ln y = \ln a + n \ln x$ . Here, both variables are logarithmic; the gradient represents the power $n$ , and the vertical intercept represents $\ln a$ .

4. Key Distinctions

It is critical to distinguish between Exponential Models and Power Models based on which variable is being logged in the linear transformation.

Feature	Exponential Model ( $y = Ab^x$ )	Power Model ( $y = ax^n$ )
Linear Form	$\ln y = (\ln b)x + \ln A$	$\ln y = n(\ln x) + \ln a$
X-axis Variable	$x$ (Linear)	$\ln x$ (Logarithmic)
Y-axis Variable	$\ln y$ (Logarithmic)	$\ln y$ (Logarithmic)
Gradient Meaning	$\ln(\text{base})$	Power ( $n$ )

Another distinction lies in Growth vs. Decay: in the form $y = Ae^{kt}$ , growth occurs when $k$ is positive, while decay occurs when $k$ is negative. In the form $y = Ab^x$ , growth occurs when $b > 1$ , and decay occurs when $0 < b < 1$ .

5. Exam Strategy & Tips

Interpret the Intercept: Always remember that the intercept on a log-linear graph is the log of the initial value, not the initial value itself. To find $A$ from a $\ln y$ intercept of $c$ , you must calculate $A = e^c$ .
Check for Asymptotes: Exponential models $y = Ae^{kt}$ never actually reach zero; they approach the x-axis asymptotically. If a model includes a vertical shift, such as $y = Ae^{kt} + C$ , the horizontal asymptote shifts to $y = C$ .
Sanity Checks: Evaluate if the model's predictions are realistic for the context. For example, a population model that predicts infinite growth is likely only valid for a short time frame before resources become a limiting factor.
Exact Values: When calculating the rate constant $k$ , keep the value in its logarithmic form (e.g., $k = \frac{1}{2} \ln 0.5$ ) until the final step to avoid rounding errors that compound significantly in exponential functions.

It is critical to distinguish between Exponential Models and Power Models based on which variable is being logged in the linear transformation.

Feature	Exponential Model ( $y = Ab^x$ )	Power Model ( $y = ax^n$ )
Linear Form	$\ln y = (\ln b)x + \ln A$	$\ln y = n(\ln x) + \ln a$
X-axis Variable	$x$ (Linear)	$\ln x$ (Logarithmic)
Y-axis Variable	$\ln y$ (Logarithmic)	$\ln y$ (Logarithmic)
Gradient Meaning	$\ln(\text{base})$	Power ( $n$ )

Interpret the Intercept: Always remember that the intercept on a log-linear graph is the log of the initial value, not the initial value itself. To find $A$ from a $\ln y$ intercept of $c$ , you must calculate $A = e^c$ .
Check for Asymptotes: Exponential models $y = Ae^{kt}$ never actually reach zero; they approach the x-axis asymptotically. If a model includes a vertical shift, such as $y = Ae^{kt} + C$ , the horizontal asymptote shifts to $y = C$ .
Sanity Checks: Evaluate if the model's predictions are realistic for the context. For example, a population model that predicts infinite growth is likely only valid for a short time frame before resources become a limiting factor.
Exact Values: When calculating the rate constant $k$ , keep the value in its logarithmic form (e.g., $k = \frac{1}{2} \ln 0.5$ ) until the final step to avoid rounding errors that compound significantly in exponential functions.