ECON0006 Chap.6 Dynamic Equations and Linear Systems
Dynamic Equations and Linear Systems
Dynamic Equations and Linear Systems follows states through discrete or continuous time. A law of motion, initial information and time convention determine a path. Stationary states reproduce themselves, while stability asks whether nearby paths return after a disturbance. Scalar equations extend to interacting systems and higher-order relations.
Exact and numerical solutions must use the same parameters, timing and initial data before their trajectories can be compared.
What this chapter covers
- 01
Discrete-time states
- 02
Continuous-time change
- 03
Linear difference equations
- 04
Linear differential equations
- 05
Stationary states
- 06
Local stability
- 07
Exact and numerical paths
Worked example · free
Separate stationarity from stability
- 1Verify the state in the no-change condition.
- 2Perturb the state above and below the candidate.
- 1Classify whether the distance shrinks or grows.
Key terms
- Law of motion
- A rule specifying how a state changes through time.
- Stationary state
- A state that reproduces itself under the law of motion.
- Local stability
- The tendency of nearby paths to return toward a reference state.
Dynamic Equations and Linear Systems FAQ
What is the difference between difference and differential equations?
Difference equations advance a state across discrete periods, while differential equations describe an instantaneous rate of change in continuous time. Their notation and data requirements must remain distinct.
Does a stationary state have to be stable?
No. It may reproduce itself exactly while nearby paths move away. Stability requires a separate perturbation or path analysis that tracks the direction of small deviations over time.
When is a numerical solution useful?
It can explore systems without a convenient exact expression, but its time step, initial data, horizon and approximation status must be reported and checked. Compare alternative step sizes when accuracy matters.
How should a time-path graph be checked?
Calculate several early updates directly from the law of motion and compare them with the plotted direction. Label the initial state and stationary candidate, inspect distance as well as sign and vary the scale or horizon when movement looks flat. This links the visible trajectory to the model’s timing and helps distinguish slow divergence, damped oscillation and genuine convergence.
Exam move
Write the time convention and state variables before copying a dynamic equation. Build a small trajectory table to see how the rule updates the state, then solve the stationary condition separately. Test nearby initial values and vary the plotting range so slow divergence or oscillation is not hidden. For systems, keep component order fixed and identify feedback from one state to another.
Compare exact and numerical paths only after matching parameters and initial data. Review errors by shifted time index, stationary state called stable, missing initial condition and approximate path reported as exact. Create a timing dictionary before manipulating a model. State whether time is discrete or continuous, what the state contains and which values are predetermined when the next movement is calculated.
For a recurrence, place next-period quantities on the left and current states and announced inputs on the right whenever that is the model’s convention. For a differential equation, distinguish a level from its instantaneous rate. Then calculate several early updates by hand. A short trajectory table catches shifted indices, omitted initial information and sign errors before they enter a general solution.
Treat stationarity and stability as separate exercises. First impose no change and verify every stationary candidate in the original rule. Next perturb above, below or along different directions and observe whether the distance from the candidate contracts. In a system, a single direction may return while another moves away, so keep the state-vector order fixed and test the feedback across components.
Describe oscillation using both alternating direction and changing distance. A path can cross the stationary value repeatedly while converging, remaining at constant amplitude or diverging, and those cases require different labels. Compare exact and numerical methods through matched inputs.
Use the same parameter values, initial data and time horizon, then report the numerical step convention and whether refinement changes the result. Inspect residuals or the original evolution rule rather than relying only on a smooth-looking plot. Vary the vertical scale so slow growth is not mistaken for a flat path, and extend the horizon when convergence is unusually slow.
If an approximation is used, label it and state the region or time span over which the comparison supports it. For exam revision, pair every calculation with a one-sentence economic trajectory. Identify the state, the shock or initial displacement, the mechanism governing adjustment and the long-run or finite-horizon implication.
Do not claim convergence from the stationary equation alone, and do not carry a stability label across a parameter change without repeating the local test. End with an audit of timing, required initial conditions, stationary candidates, path classification and exact-versus-numerical status. This sequence keeps the dynamic argument readable even if the final closed form is unavailable.
Working through Dynamic Equations and Linear Systems in ECON0006? Sia is AskSia’s AI Economics tutor — ask any ECON0006 Dynamic Equations and Linear Systems question and get a clear, step-by-step explanation grounded in how ECON0006 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.