1.1.2. Types of Finite Differences

a. Forward Difference (Δ) Δ𝑓(𝑥)= 𝑓(𝑥+ ℎ)− 𝑓(𝑥) Higher order forward differences: Δ²𝑓(𝑥) = Δ(Δ𝑓(𝑥))= 𝑓(𝑥 +2ℎ)− 2𝑓(𝑥+ ℎ)+ 𝑓(𝑥) Δ³𝑓(𝑥) = Δ(Δ²𝑓(𝑥)), and so on.

The forward difference is a way of estimating the derivative of a function. Essentially, it measures how much the function changes as we move forward by a small distance, h. The first order forward difference, Δf(x), compares the function value at x+h with the function value at x. For higher orders, like Δ²f(x), we apply the forward difference process again to capture how the first difference itself changes. This is useful for approximating derivatives when you only have discrete data points.

Imagine you’re watching a car moving down a straight pathway. If you note the car's position at regular intervals (say every second), the forward difference would tell you how far the car has traveled from one position to the next. If you wanted to know how the car’s speed changes (like speeding up or slowing down), you would look at the changes in these distances over time, which represents higher-order differences.

b. Backward Difference (∇) ∇𝑓(𝑥)= 𝑓(𝑥)−𝑓(𝑥−ℎ) Higher order backward differences: ∇²𝑓(𝑥)= ∇(∇𝑓(𝑥))= 𝑓(𝑥)−2𝑓(𝑥− ℎ)+ 𝑓(𝑥− 2ℎ)

The backward difference is similar to the forward difference, but it looks at changes in the opposite direction. It calculates the difference between the function value at a point x and its value a little before it, at x-h. This approach can also be extended to higher orders, where you consider how the differences themselves change. This method is useful when your data points come from the past and you're interested in trends leading up to the current moment.

Think of a sports game replay. If you’re watching a football match, and you want to analyze how a player performed just before scoring a goal, you would look at their past movements – that’s what the backward difference does! It helps in understanding how previous actions lead to the current situation.

c. Central Difference (𝛿) ℎ ℎ 𝛿𝑓(𝑥) = 𝑓(𝑥+ ) −𝑓(𝑥− ) 2 2 It is a symmetric difference and gives better accuracy in approximating derivatives.

The central difference takes into account values of the function at both sides of a point, instead of just looking forward or backward. By averaging the changes from both directions, it tends to provide a more accurate approximation of the derivative at x, especially when dealing with smooth functions. This balance can result in lower error in many practical applications.

Imagine tasting a dish to figure out if it’s too salty. If you taste from the left side of the dish (forward) and the right side (backward), then average those flavors, you probably get a better sense of the overall taste than if you only sampled one side. That’s how the central difference captures a more accurate picture of the function's behavior.

d. Shift Operator (𝐸) 𝐸𝑓(𝑥) = 𝑓(𝑥+ℎ) Using 𝐸, we define: Δ = 𝐸− 1, ∇= 1 −𝐸−1

The shift operator E is a convenient mathematically abstract way to notate shifts in function arguments. Specifically, applying E to a function f(x) shifts it to the right by h. The forward and backward differences can then be expressed in terms of E. This can simplify computations, especially when analyzing sequences of differences.

Think of a broadcasting radio. When you change the dial a certain amount (let’s say ‘h’), you’re shifting to a new frequency. The shift operator does something similar with function values. It allows you to shift your focus and analyze how a function behaves as you move along its inputs.

e. Average Operator (𝜇) ℎ ℎ 𝑓(𝑥 + ) +𝑓(𝑥− ) 2 2 𝜇𝑓(𝑥) = 2 This is often used in central difference formulas.

The average operator computes the average of the function values at points just before and after x. This averaging is particularly helpful in central differences as it smooths out fluctuations, giving a clearer estimate of the function's behavior at that point. It effectively provides symmetry and stability to the difference calculations.

If you're measuring a person's height by taking an average from two measuring tapes, one on each side, you’d get a more accurate height than measuring with just one tape. That’s what the average operator does – it balances the input to give a clearer outcome.

--