Using Exponentials in Modelling

The Initial Value Parameter ($A$) represents the state of the system at the start of the observation ($t = 0$). Because $e^0 = 1$, the equation simplifies to $y = A(1)$, identifying $A$ as the y-intercept on a Cartesian plane.

The Growth/Decay Constant ($k$) determines the speed and direction of the change. A positive $k$ indicates that the quantity is increasing at an accelerating rate, while a negative $k$ indicates the quantity is approaching zero asymptotically.

Proportional Growth is the logical foundation of these models; the rate of change $\frac{dy}{dt}$ is directly proportional to $y$. This means as the population or value gets larger, it grows (or shrinks) faster in absolute terms.

Identifying the Model Type: The first step is determining if the scenario describes growth (e.g., interest, bacteria) or decay (e.g., cooling, depreciation). This dictates whether the exponent in $y = Ae^{kt}$ should be positive or negative.

Solving for Parameters: If $A$ is not given, it is found by substituting $t=0$. To find $k$, a second known data point $(t_1, y_1)$ must be substituted into the equation, followed by the application of natural logarithms to isolate $k$.

Predicting Future Values: Once $A$ and $k$ are established, the model becomes a predictive tool. By substituting any future value of $t$, the corresponding value of $y$ can be calculated, or vice versa by solving the resulting logarithmic equation.

Linear vs. Exponential: Unlike linear models where a quantity increases by a fixed amount per unit of time, exponential models increase by a fixed percentage or ratio. This leads to much more rapid changes over long periods.

Discrete vs. Continuous: While some models use discrete steps (like annual interest), the $e^{kt}$ form assumes change is happening continuously at every infinitesimal moment.

Check the Initial Condition: Always verify if the 'initial' value provided in a problem corresponds to $t=0$. If the data starts at a different time, you must adjust your $t$ values accordingly to maintain the accuracy of the $A$ parameter.

Logarithmic Accuracy: When solving for $k$, avoid rounding intermediate values. Use the exact logarithmic expression (e.g., $k = \frac{1}{5}\ln(2)$) in your calculator for subsequent steps to prevent significant rounding errors in the final answer.

Sanity Checks: Evaluate if your calculated $k$ matches the context. If a value is supposed to be depreciating but you calculate a positive $k$, re-examine your algebraic steps for a sign error.

Units and Context: Always state the units for your final answer (e.g., 'dollars', 'grams', 'years'). Examiners often look for the interpretation of the mathematical result within the real-world context provided.

The 'Zero' Misconception: Students often assume exponential decay will eventually reach zero. Mathematically, $Ae^{-kt}$ never reaches zero; it only gets infinitely close, which is a critical distinction when discussing 'total' depletion.

Confusing $A$ with $y$: A common error is substituting the initial value for $y$ instead of $A$. Remember that $A$ is a constant for the specific model, while $y$ is the variable that changes as $t$ progresses.

Incorrect Log Application: When solving $y = Ae^{kt}$ for $t$, students sometimes try to take the log before dividing by $A$. You must isolate the exponential term first: $\frac{y}{A} = e^{kt} \implies \ln(\frac{y}{A}) = kt$.