OctoBot 1.0.2: ChatGPT Backtesting, DCA, and TradingView Updates
Summary
This release note describes changes in OctoBot 1.0.2. ChatGPT strategy profiles can now be backtested for some settings using prompts precomputed from historical exchange-pair data downloaded by the service. The ChatGPT profile also shifts from daily trading to a DCA mode intended to shape sell orders after an entry signal. Users can configure prompts to use candle history without technical indicators and choose the number of candles included.
TradingView integration gains alert signals to cancel all open orders for a symbol or cancel orders on a specified side. The release also reports a backtesting-engine bug fix for orders that could fill prematurely. These are product changes, not evidence that ChatGPT signals or DCA produce profitable outcomes. Backtesting availability is limited to some settings, and the note gives no evaluation results for the revised strategy or details on the bug fix's scope.
Key ideas
- Some ChatGPT settings can be backtested using prompts precomputed from historical exchange-pair data.\nThe ChatGPT profile changes from daily trading to DCA after an entry signal.\nPrompt settings can use candle history without technical indicators and specify a candle count.\nTradingView alerts can cancel all open orders for a symbol or orders on a selected side.\nThe release fixes premature fills in the backtesting engine but reports no strategy performance results.
Tags
Full text
# OctoBot 1.0.2 - What's new --- title: "OctoBot 1.0.2 - What's new" slug: "octobot-1-0-2-whats-new" date: "2023-10-30" authors: ["guillaume"] tags: ["Tradingview", "Chatgpt", "Release", "DCA", "Backtesting"] image: "/images/blog/octobot-1-0-2-whats-new/cover.png" --- # OctoBot 1.0.2 - What's new  ## Introducing OctoBot 1.0.2 We're thrilled to announce the release of OctoBot 1.0.2, an upgraded version with many improved features, thanks to the great feedback we received from you all. ## Revamped ChatGPT strategy In OctoBot 1.0.2, we've revamped the ChatGPT strategy. Until now, you couldn't run a [backtesting](/guides/octobot-usage/backtesting) on a chatgpt profile due to the excessive prompt, costing around $2 for 6 months history, hence we disallowed it. However, with the new update, you can run backtesting on some gpt settings because we've already computed the prompt against some exchanges pairs historical data which are downloaded from our servers. We've also shifted from Daily Trading mode to a smart DCA trading mode in the chatgpt profile. The previous mode was no longer suited to the current market, hence we updated it to DCA trading mode to develop more accurate sell orders following a chatgpt entry signal. Additionally, we've introduced a new prompt setting. You can now ask chatgpt with pure candle history (without any TA indicator) and include the number of candles you want.  ## Improved TradingView connection We've also made noteworthy improvements to the TradingView connection, thanks to some valuable feedback from our OctoBot users who use the TradingView integration. It's now possible to send a cancel order signal to cancel all current open orders for a symbol, or only to cancel an open order on a specific side using the param SIDE. More details on this can be found on [the alert format guide](/guides/octobot-interfaces/tradingview/alert-format#canceling-orders). Special thanks to @KidCharlemagne, an active member of our OctoBot <a href="https://discord.com/invite/vHkcb8W" rel="nofollow">Discord community</a>, for helping with the complete refactor of the TradingView [configuration guide](/guides/octobot-interfaces/tradingview). It's clearer now, with ample examples.  ## Some bugfix We've also squashed some bugs in this release. After careful checks, we discovered an issue in the OctoBot [backtesting engine](/guides/octobot-usage/backtesting) that allowed for premature filling of open orders. ## Conclusion We can't wait to hear your thoughts on this new version. Please use this <a href="https://feedback.octobot.cloud/open-source" rel="nofollow">feedback link</a> to share your suggestions and what you'd like to see in our next release.
Shown in full with attribution under the source's licence. Licence: GPL-3.0
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.