Constructing Euler Inputs from an Initial-Value Problem

From an Initial-Value Problem to Its Inputs

A first-order differential equation describes how a dependent variable changes as an independent variable changes. An initial-value problem adds a specified value of the dependent variable at a particular independent-variable value. Before applying a numerical method, these pieces must be identified clearly.

For forward Euler, the relevant information is the starting point, the starting value, and a function that gives the slope at any point. The equation must be expressed so that the derivative is isolated. This tutorial focuses on recognizing and constructing those inputs; it does not calculate successive numerical approximations.

When the independent variable represents time, it is common to write \(t\) instead of \(x\), and \(y'(t)\) or \(\frac{dy}{dt}\) for the derivative. The same structure then reads \(y'=f(t,y)\), \(y(t_0)=y_0\). The letters can change, but the roles do not.

Recognizing Each Part

The independent variable is the quantity with respect to which the derivative is taken. In \(\frac{dy}{dt}\), it is \(t\); in \(\frac{dy}{dx}\), it is \(x\). The dependent variable is the quantity whose value changes as the independent variable changes. In these examples it is \(y\), though models often use names such as \(P\) for population or \(C\) for concentration.

The initial condition specifies the dependent variable’s value at one independent-variable value. For example, \(y(2)=5\) says that the initial point is \(x_0=2\) and the initial value is \(y_0=5\). The independent variable and its initial value must be read together: \(2\) is the starting input, while \(5\) is the starting output.

The slope function is the right-hand side of the differential equation once the derivative is isolated. If \(y'=f(x,y)\), then at a point \((x,y)\), the derivative—and therefore the slope prescribed by the equation—is \(f(x,y)\). The function may depend on \(x\), on \(y\), or on both.

Why Explicit Form Gives the Slope

A solution to the initial-value problem is a differentiable function \(y(x)\) that satisfies both the differential equation and the initial condition. The differential equation says that, at every point on that solution, its derivative equals the value of \(f\) at the current coordinates. In particular, at the initial point, the slope is determined by substituting the initial coordinates into \(f\).

This follows directly from the two defining conditions: the differential equation gives \(y'(x)=f(x,y(x))\), and the initial condition gives \(y(x_0)=y_0\). Substitution of \(x=x_0\) into the first equality therefore gives the displayed result. It explains why the initial point and slope function together determine the starting slope, without requiring us to solve the differential equation.

Units are also a useful check. If \(x\) is measured in days and \(y\) in grams, then \(dy/dx\) and the values of \(f\) must have units of grams per day. A mismatch can reveal an incorrect rate conversion or a mistaken interpretation of the variables.

Worked Examples

Conditions and Careful Interpretation

An equation is in explicit first-order form only when it specifies the first derivative as a function of the independent and dependent variables. For instance, an equation that contains \((1+y^2)y'\) is not yet written in that form, even if it can be rearranged into it. Before dividing by an expression, check whether it can be zero. If it can, division may exclude points or branches where the original equation was meaningful; the rearranged equation would then need a stated domain restriction.

A first-order initial-value problem normally specifies one initial value for its dependent variable. The condition \(y(x_0)=y_0\) gives one point on a solution. It does not state the slope directly; the slope at that point is found from \(f(x_0,y_0)\). Nor does the condition alone provide the slope function: that comes from the differential equation.

When the slope function depends only on the dependent variable, as in \(P'=0.08P\), it is called autonomous. When it depends explicitly on the independent variable as well, it is nonautonomous. For example, \(y'=x-y\) depends on both \(x\) and \(y\). This distinction is useful for describing the equation, but both cases use the same input format \(y'=f(x,y)\), \(y(x_0)=y_0\).

Common Mistake: Swapping the Initial Coordinates

A frequent error is to read \(y(1)=2\) as \(x_0=2\), \(y_0=1\). That reverses the input and output: the expression \(y(1)\) means the value of \(y\) when the input is \(1\). The correction is \(x_0=1\), \(y_0=2\).

For the equation in the preceding example, this mistake also changes the calculated initial slope. The incorrect substitution would produce

whereas the correct substitution is

These values differ in both sign and magnitude. To avoid this error, translate \(y(x_0)=y_0\) into words: “when the independent variable is \(x_0\), the dependent variable is \(y_0\).” Then substitute the pair in that order into the slope function.

A Reliable Input-Extraction Process

These steps separate the information supplied by the initial condition from the information supplied by the differential equation. The initial condition fixes the starting point; the slope function describes the derivative rule. Together they put a first-order IVP into the form needed for a forward Euler calculation. A step size is a separate numerical choice, not part of the IVP’s initial data.