|
|
|
@ -1,6 +1,6 @@ |
|
|
|
# X-Ray Feature Notes — What Exists Today |
|
|
|
|
|
|
|
**Author:** Sesha Siva Sankar (Member 1) |
|
|
|
**Author:** sesha siva sankar (member 1) |
|
|
|
**Date:** Tue, Sep 15 |
|
|
|
**Task:** Read existing Ghostfolio X-Ray code and document the current rule-based logic |
|
|
|
|
|
|
|
@ -48,35 +48,35 @@ all fits together, explained the way I'd explain it to a teammate over coffee, n |
|
|
|
Here's the full list, grouped the way the app groups them: |
|
|
|
|
|
|
|
- **Liquidity** |
|
|
|
- *Buying Power* — checks if your uninvested cash is below a minimum threshold (default: |
|
|
|
- _Buying Power_ — checks if your uninvested cash is below a minimum threshold (default: |
|
|
|
just checks it's not zero / not too low). Fully configurable. |
|
|
|
|
|
|
|
- **Emergency Fund** |
|
|
|
- *Setup* — checks if you've even set an emergency fund target. No threshold, just |
|
|
|
- _Setup_ — checks if you've even set an emergency fund target. No threshold, just |
|
|
|
"did you set one or not." |
|
|
|
- *Coverage* — checks if your tagged emergency-fund holdings + cash actually cover your |
|
|
|
- _Coverage_ — checks if your tagged emergency-fund holdings + cash actually cover your |
|
|
|
target. Can fail three ways: no target set, holdings alone overshoot the target |
|
|
|
(weird case), or holdings+cash undershoot it. Only runs if you've set a target > 0. |
|
|
|
|
|
|
|
- **Currency Cluster Risk** (are you too concentrated in one currency?) |
|
|
|
- *Current Investment* — fails if your single biggest currency is more than 50% of your |
|
|
|
- _Current Investment_ — fails if your single biggest currency is more than 50% of your |
|
|
|
portfolio value (configurable %). |
|
|
|
- *Base Currency Investment* — fails if your biggest currency bucket (by money invested) |
|
|
|
- _Base Currency Investment_ — fails if your biggest currency bucket (by money invested) |
|
|
|
isn't your home currency. No threshold to tune, it's just yes/no. |
|
|
|
|
|
|
|
- **Asset Class Cluster Risk** (stocks vs bonds balance) |
|
|
|
- *Equity* — fails if equities are outside a 78%–82% band. |
|
|
|
- *Fixed Income* — fails if bonds are outside an 18%–22% band. |
|
|
|
- Interesting detail: these are *band* checks (too high OR too low both fail), not simple |
|
|
|
- _Equity_ — fails if equities are outside a 78%–82% band. |
|
|
|
- _Fixed Income_ — fails if bonds are outside an 18%–22% band. |
|
|
|
- Interesting detail: these are _band_ checks (too high OR too low both fail), not simple |
|
|
|
ceilings like most of the others. |
|
|
|
|
|
|
|
- **Account Cluster Risk** |
|
|
|
- *Current Investment* — fails if one account holds more than 50% of your total money. |
|
|
|
- *Single Account* — fails if you only have one account at all (or zero, which is invalid). |
|
|
|
- _Current Investment_ — fails if one account holds more than 50% of your total money. |
|
|
|
- _Single Account_ — fails if you only have one account at all (or zero, which is invalid). |
|
|
|
|
|
|
|
- **Economic Market Cluster Risk** (developed vs emerging market split) |
|
|
|
- *Developed Markets* — band check, must be within 68%–72%. |
|
|
|
- *Emerging Markets* — band check, must be within 28%–32%. |
|
|
|
- _Developed Markets_ — band check, must be within 68%–72%. |
|
|
|
- _Emerging Markets_ — band check, must be within 28%–32%. |
|
|
|
|
|
|
|
- **Regional Market Cluster Risk** (geographic spread) |
|
|
|
- Five separate rules, one per region: Asia Pacific, Emerging Markets, Europe, Japan, |
|
|
|
@ -84,7 +84,7 @@ Here's the full list, grouped the way the app groups them: |
|
|
|
(e.g. North America should be 65%–69% of your advanced-market exposure). |
|
|
|
|
|
|
|
- **Fees** |
|
|
|
- *Fee Ratio* — fails if total fees paid are more than 1% of your total trading volume |
|
|
|
- _Fee Ratio_ — fails if total fees paid are more than 1% of your total trading volume |
|
|
|
(buys + sells combined). Configurable. |
|
|
|
|
|
|
|
## 4. Can users customize any of this? |
|
|
|
|